Merge pull request #2972 from xinnan-tech/update-tts-voice-data

Update tts voice data

.gitignore

@@ -3,6 +3,9 @@ __pycache__/
 .idea/
 *.py[cod]
 *$py.class
+.vscode
+.claude
+AGENTS.md
 
 # C extensions
 *.so

main/manager-api/src/main/java/xiaozhi/modules/agent/dto/AgentUpdateDTO.java

@@ -1,6 +1,7 @@
 package xiaozhi.modules.agent.dto;
 
 import java.io.Serializable;
+import java.math.BigDecimal;
 import java.util.HashMap;
 import java.util.List;
 
@@ -41,6 +42,18 @@ public class AgentUpdateDTO implements Serializable {
     @Schema(description = "音色标识", example = "voice_02", nullable = true)
     private String ttsVoiceId;
 
+    @Schema(description = "音色语言", example = "普通话", nullable = true)
+    private String ttsLanguage;
+
+    @Schema(description = "TTS音量", example = "50", nullable = true)
+    private Integer ttsVolume;
+
+    @Schema(description = "TTS语速", example = "50", nullable = true)
+    private Integer ttsRate;
+
+    @Schema(description = "TTS音调", example = "50", nullable = true)
+    private Integer ttsPitch;
+
     @Schema(description = "记忆模型标识", example = "mem_model_02", nullable = true)
     private String memModelId;
 

main/manager-api/src/main/java/xiaozhi/modules/agent/entity/AgentEntity.java

@@ -1,5 +1,6 @@
 package xiaozhi.modules.agent.entity;
 
+import java.math.BigDecimal;
 import java.util.Date;
 
 import com.baomidou.mybatisplus.annotation.IdType;
@@ -45,6 +46,18 @@ public class AgentEntity {
     @Schema(description = "音色标识")
     private String ttsVoiceId;
 
+    @Schema(description = "音色语言")
+    private String ttsLanguage;
+
+    @Schema(description = "TTS音量")
+    private Integer ttsVolume;
+
+    @Schema(description = "TTS语速")
+    private Integer ttsRate;
+
+    @Schema(description = "TTS音调")
+    private Integer ttsPitch;
+
     @Schema(description = "记忆模型标识")
     private String memModelId;
 

main/manager-api/src/main/java/xiaozhi/modules/agent/entity/AgentTemplateEntity.java

@@ -1,6 +1,7 @@
 package xiaozhi.modules.agent.entity;
 
 import java.io.Serializable;
+import java.math.BigDecimal;
 import java.util.Date;
 
 import com.baomidou.mybatisplus.annotation.IdType;
@@ -64,6 +65,26 @@ public class AgentTemplateEntity implements Serializable {
      */
     private String ttsVoiceId;
 
+    /**
+     * 音色语言
+     */
+    private String ttsLanguage;
+
+    /**
+     * TTS音量
+     */
+    private Integer ttsVolume;
+
+    /**
+     * TTS语速
+     */
+    private Integer ttsRate;
+
+    /**
+     * TTS音调
+     */
+    private Integer ttsPitch;
+
     /**
      * 记忆模型标识
      */

main/manager-api/src/main/java/xiaozhi/modules/agent/service/impl/AgentServiceImpl.java

@@ -309,6 +309,18 @@ public class AgentServiceImpl extends BaseServiceImpl<AgentDao, AgentEntity> imp
         if (dto.getTtsVoiceId() != null) {
             existingEntity.setTtsVoiceId(dto.getTtsVoiceId());
         }
+        if (dto.getTtsLanguage() != null) {
+            existingEntity.setTtsLanguage(dto.getTtsLanguage());
+        }
+        if (dto.getTtsVolume() != null) {
+            existingEntity.setTtsVolume(dto.getTtsVolume());
+        }
+        if (dto.getTtsRate() != null) {
+            existingEntity.setTtsRate(dto.getTtsRate());
+        }
+        if (dto.getTtsPitch() != null) {
+            existingEntity.setTtsPitch(dto.getTtsPitch());
+        }
         if (dto.getMemModelId() != null) {
             existingEntity.setMemModelId(dto.getMemModelId());
         }

main/manager-api/src/main/java/xiaozhi/modules/config/service/impl/ConfigServiceImpl.java

@@ -87,6 +87,10 @@ public class ConfigServiceImpl implements ConfigService {
                 null,
                 null,
                 null,
+                null,
+                null,
+                null,
+                null,
                 agent.getVadModelId(),
                 agent.getAsrModelId(),
                 null,
@@ -135,15 +139,24 @@ public class ConfigServiceImpl implements ConfigService {
         String voice = null;
         String referenceAudio = null;
         String referenceText = null;
+        String language = null;
         TimbreDetailsVO timbre = timbreService.get(agent.getTtsVoiceId());
         if (timbre != null) {
             voice = timbre.getTtsVoice();
             referenceAudio = timbre.getReferenceAudio();
             referenceText = timbre.getReferenceText();
+            // 优先使用用户选择的语言，如果没有则使用音色支持的第一个语言
+            if (StringUtils.isNotBlank(agent.getTtsLanguage())) {
+                language = agent.getTtsLanguage();
+            } else if (StringUtils.isNotBlank(timbre.getLanguages())) {
+                language = timbre.getLanguages().split("、")[0].trim();
+            }
         } else {
             VoiceCloneEntity voice_print = cloneVoiceService.selectById(agent.getTtsVoiceId());
             if (voice_print != null) {
                 voice = voice_print.getVoiceId();
+                // 优先使用用户选择的语言，如果没有则使用默认值
+                language = StringUtils.isNotBlank(agent.getTtsLanguage()) ? agent.getTtsLanguage() : "普通话";
             }
         }
         // 构建返回数据
@@ -208,6 +221,10 @@ public class ConfigServiceImpl implements ConfigService {
                 voice,
                 referenceAudio,
                 referenceText,
+                language,
+                agent.getTtsVolume(),
+                agent.getTtsRate(),
+                agent.getTtsPitch(),
                 agent.getVadModelId(),
                 agent.getAsrModelId(),
                 agent.getLlmModelId(),
@@ -385,6 +402,10 @@ public class ConfigServiceImpl implements ConfigService {
             String voice,
             String referenceAudio,
             String referenceText,
+            String language,
+            Integer ttsVolume,
+            Integer ttsRate,
+            Integer ttsPitch,
             String vadModelId,
             String asrModelId,
             String llmModelId,
@@ -423,6 +444,14 @@ public class ConfigServiceImpl implements ConfigService {
                         ((Map<String, Object>) model.getConfigJson()).put("ref_audio", referenceAudio);
                     if (referenceText != null)
                         ((Map<String, Object>) model.getConfigJson()).put("ref_text", referenceText);
+                    if (language != null)
+                        ((Map<String, Object>) model.getConfigJson()).put("language", language);
+                    if (ttsVolume != null)
+                        ((Map<String, Object>) model.getConfigJson()).put("ttsVolume", ttsVolume);
+                    if (ttsRate != null)
+                        ((Map<String, Object>) model.getConfigJson()).put("ttsRate", ttsRate);
+                    if (ttsPitch != null)
+                        ((Map<String, Object>) model.getConfigJson()).put("ttsPitch", ttsPitch);
 
                     // 火山引擎声音克隆需要替换resource_id
                     Map<String, Object> map = (Map<String, Object>) model.getConfigJson();

main/manager-api/src/main/java/xiaozhi/modules/model/dto/VoiceDTO.java

@@ -23,6 +23,9 @@ public class VoiceDTO implements Serializable {
     @Schema(description = "音频播放地址")
     private String voiceDemo;
     
+    @Schema(description = "语言类型")
+    private String languages;
+    
     @Schema(description = "是否为克隆音色")
     private Boolean isClone;
 
@@ -31,6 +34,7 @@ public class VoiceDTO implements Serializable {
         this.id = id;
         this.name = name;
         this.voiceDemo = null;
+        this.languages = null;
         this.isClone = false; // 默认不是克隆音色
     }
     
@@ -39,6 +43,7 @@ public class VoiceDTO implements Serializable {
         this.id = id;
         this.name = name;
         this.voiceDemo = voiceDemo;
+        this.languages = null;
         this.isClone = false;
     }
 

main/manager-api/src/main/java/xiaozhi/modules/timbre/service/impl/TimbreServiceImpl.java

@@ -129,6 +129,7 @@ public class TimbreServiceImpl extends BaseServiceImpl<TimbreDao, TimbreEntity>
                 .map(entity -> {
                     VoiceDTO dto = new VoiceDTO(entity.getId(), entity.getName());
                     dto.setVoiceDemo(entity.getVoiceDemo());
+                    dto.setLanguages(entity.getLanguages()); // 设置语言类型
                     dto.setIsClone(false); // 设置为普通音色
                     return dto;
                 })
@@ -146,6 +147,7 @@ public class TimbreServiceImpl extends BaseServiceImpl<TimbreDao, TimbreEntity>
                 voiceDTO.setName(MessageUtils.getMessage(ErrorCode.VOICE_CLONE_PREFIX) + entity.getName());
                 // 保留从数据库查询到的voiceDemo字段
                 voiceDTO.setVoiceDemo(entity.getVoiceDemo());
+                voiceDTO.setLanguages(entity.getLanguages());
                 voiceDTO.setIsClone(true); // 设置为克隆音色
                 redisUtils.set(RedisKeys.getTimbreNameById(voiceDTO.getId()), voiceDTO.getName(),
                         RedisUtils.NOT_EXPIRE);

main/manager-api/src/main/resources/db/changelog/202602061650.sql

@@ -0,0 +1,48 @@
+-- 统一规范ai_tts_voice语言类型数据
+UPDATE ai_tts_voice
+SET languages = CASE
+    WHEN languages IN ('中文', '普通话','东北话','天津话','中文-北京口音','中文-青岛口音','中文-河南口音','中文-广西口音','辽宁','陕西','中文-四川口音','中文-台湾口音','中文-长沙口音') THEN '普通话'
+    WHEN languages IN ('中文及中英文混合', '中文、英文', '中文、美式英语','中文-北京口音、英文','中文(东北)及中英文混合') THEN '普通话、英语'
+    WHEN languages IN ('英式英文', '英式英语', '美式英语', '澳洲英语', '英文') THEN '英语'
+    WHEN languages = '日语' THEN '日语'
+    WHEN languages = '日语、西语' THEN '日语、西班牙语'
+    WHEN languages = '韩语' THEN '韩语'
+    WHEN languages IN ('粤语', '中文-广东口音') THEN '粤语'
+    WHEN languages = '中文(粤语)及中英文混合' THEN '粤语、英语'
+    WHEN languages = '粤语及粤英混合' THEN '粤语、英语'
+    ELSE languages
+END;
+
+-- 添加音色语言、音量、语速、音调字段到 ai_agent 表
+SET @col_exists = (SELECT COUNT(*) FROM INFORMATION_SCHEMA.COLUMNS WHERE TABLE_SCHEMA = DATABASE() AND TABLE_NAME = 'ai_agent' AND COLUMN_NAME = 'tts_language');
+SET @sql = IF(@col_exists = 0, 'ALTER TABLE `ai_agent` ADD COLUMN `tts_language` VARCHAR(50) NULL COMMENT ''音色语言'' AFTER `tts_voice_id`', 'SELECT ''Column tts_language already exists'' AS msg');
+PREPARE stmt FROM @sql; EXECUTE stmt; DEALLOCATE PREPARE stmt;
+
+SET @col_exists = (SELECT COUNT(*) FROM INFORMATION_SCHEMA.COLUMNS WHERE TABLE_SCHEMA = DATABASE() AND TABLE_NAME = 'ai_agent' AND COLUMN_NAME = 'tts_volume');
+SET @sql = IF(@col_exists = 0, 'ALTER TABLE `ai_agent` ADD COLUMN `tts_volume` INT NULL COMMENT ''TTS音量'' AFTER `tts_language`', 'SELECT ''Column tts_volume already exists'' AS msg');
+PREPARE stmt FROM @sql; EXECUTE stmt; DEALLOCATE PREPARE stmt;
+
+SET @col_exists = (SELECT COUNT(*) FROM INFORMATION_SCHEMA.COLUMNS WHERE TABLE_SCHEMA = DATABASE() AND TABLE_NAME = 'ai_agent' AND COLUMN_NAME = 'tts_rate');
+SET @sql = IF(@col_exists = 0, 'ALTER TABLE `ai_agent` ADD COLUMN `tts_rate` INT NULL COMMENT ''TTS语速'' AFTER `tts_volume`', 'SELECT ''Column tts_rate already exists'' AS msg');
+PREPARE stmt FROM @sql; EXECUTE stmt; DEALLOCATE PREPARE stmt;
+
+SET @col_exists = (SELECT COUNT(*) FROM INFORMATION_SCHEMA.COLUMNS WHERE TABLE_SCHEMA = DATABASE() AND TABLE_NAME = 'ai_agent' AND COLUMN_NAME = 'tts_pitch');
+SET @sql = IF(@col_exists = 0, 'ALTER TABLE `ai_agent` ADD COLUMN `tts_pitch` INT NULL COMMENT ''TTS音调'' AFTER `tts_rate`', 'SELECT ''Column tts_pitch already exists'' AS msg');
+PREPARE stmt FROM @sql; EXECUTE stmt; DEALLOCATE PREPARE stmt;
+
+-- 添加音色语言、音量、语速、音调字段到 ai_agent_template 表
+SET @col_exists = (SELECT COUNT(*) FROM INFORMATION_SCHEMA.COLUMNS WHERE TABLE_SCHEMA = DATABASE() AND TABLE_NAME = 'ai_agent_template' AND COLUMN_NAME = 'tts_language');
+SET @sql = IF(@col_exists = 0, 'ALTER TABLE `ai_agent_template` ADD COLUMN `tts_language` VARCHAR(50) NULL COMMENT ''音色语言'' AFTER `tts_voice_id`', 'SELECT ''Column tts_language already exists'' AS msg');
+PREPARE stmt FROM @sql; EXECUTE stmt; DEALLOCATE PREPARE stmt;
+
+SET @col_exists = (SELECT COUNT(*) FROM INFORMATION_SCHEMA.COLUMNS WHERE TABLE_SCHEMA = DATABASE() AND TABLE_NAME = 'ai_agent_template' AND COLUMN_NAME = 'tts_volume');
+SET @sql = IF(@col_exists = 0, 'ALTER TABLE `ai_agent_template` ADD COLUMN `tts_volume` INT NULL COMMENT ''TTS音量'' AFTER `tts_language`', 'SELECT ''Column tts_volume already exists'' AS msg');
+PREPARE stmt FROM @sql; EXECUTE stmt; DEALLOCATE PREPARE stmt;
+
+SET @col_exists = (SELECT COUNT(*) FROM INFORMATION_SCHEMA.COLUMNS WHERE TABLE_SCHEMA = DATABASE() AND TABLE_NAME = 'ai_agent_template' AND COLUMN_NAME = 'tts_rate');
+SET @sql = IF(@col_exists = 0, 'ALTER TABLE `ai_agent_template` ADD COLUMN `tts_rate` INT NULL COMMENT ''TTS语速'' AFTER `tts_volume`', 'SELECT ''Column tts_rate already exists'' AS msg');
+PREPARE stmt FROM @sql; EXECUTE stmt; DEALLOCATE PREPARE stmt;
+
+SET @col_exists = (SELECT COUNT(*) FROM INFORMATION_SCHEMA.COLUMNS WHERE TABLE_SCHEMA = DATABASE() AND TABLE_NAME = 'ai_agent_template' AND COLUMN_NAME = 'tts_pitch');
+SET @sql = IF(@col_exists = 0, 'ALTER TABLE `ai_agent_template` ADD COLUMN `tts_pitch` INT NULL COMMENT ''TTS音调'' AFTER `tts_rate`', 'SELECT ''Column tts_pitch already exists'' AS msg');
+PREPARE stmt FROM @sql; EXECUTE stmt; DEALLOCATE PREPARE stmt;

main/manager-api/src/main/resources/db/changelog/202602271137.sql

@@ -0,0 +1,42 @@
+-- 更新腾讯TTS供应器配置，增加speed、volume和format参数
+UPDATE `ai_model_provider`
+SET fields = '[{"key":"appid","label":"应用ID","type":"string"},{"key":"secret_id","label":"Secret ID","type":"string"},{"key":"secret_key","label":"Secret Key","type":"string"},{"key":"format","label":"音频格式","type":"string"},{"key":"speed","label":"语速","type":"number"},{"key":"volume","label":"音量","type":"number"},{"key":"output_dir","label":"输出目录","type":"string"},{"key":"voice","label":"音色ID","type":"string"},{"key":"region","label":"区域","type":"string"}]'
+WHERE id = 'SYSTEM_TTS_TencentTTS';
+
+-- 更新腾讯TTS模型配置，增加speed和volume参数，补充参数说明
+UPDATE `ai_model_config` SET 
+    `config_json` = JSON_SET(`config_json`, '$.speed', 0, '$.volume', 0),
+    `remark` = '腾讯TTS配置说明：
+1. 需要在腾讯云平台开通智能语音交互服务
+2. 支持多种音色，当前配置使用101001
+3. 需要网络连接
+4. 输出文件保存在tmp/目录
+申请步骤：
+1. 访问 https://console.cloud.tencent.com/cam/capi 获取密钥
+2. 访问 https://console.cloud.tencent.com/tts/resourcebundle 领取免费资源
+3. 创建新应用
+4. 获取appid、secret_id和secret_key
+5. 填入配置文件中
+音频参数：
+- format: 音频格式，支持pcm、wav、mp3
+- speed: 语速，范围-2~6，默认0
+- volume: 音量，范围-10~10，默认0'
+WHERE `id` = 'TTS_TencentTTS';
+
+-- 更新CozeCnTTS供应器配置，增加speed和loudness_rate参数
+UPDATE `ai_model_provider`
+SET fields = '[{"key":"voice","label":"音色","type":"string"},{"key":"access_token","label":"访问令牌","type":"string"},{"key":"speed","label":"语速","type":"number"},{"key":"loudness_rate","label":"音量增益","type":"number"},{"key":"output_dir","label":"输出目录","type":"string"},{"key":"response_format","label":"响应格式","type":"string"}]'
+WHERE id = 'SYSTEM_TTS_cozecn';
+
+-- 更新CozeCnTTS模型配置，增加speed和loudness_rate参数，补充参数说明
+UPDATE `ai_model_config` SET 
+    `config_json` = JSON_SET(`config_json`, '$.speed', 1, '$.loudness_rate', 0),
+    `remark` = 'Coze中文语音合成配置说明：
+1. 访问 https://www.coze.cn/ 注册并登录
+2. 创建应用并获取access_token
+3. 选择合适的音色ID
+音频参数：
+- response_format: 音频格式，支持pcm、wav、mp3
+- speed: 语速，范围0.5~2，默认1
+- loudness_rate: 音量增益，范围-50~100，默认0'
+WHERE `id` = 'TTS_CozeCnTTS';

main/manager-api/src/main/resources/db/changelog/db.changelog-master.yaml

@@ -530,6 +530,19 @@ databaseChangeLog:
             encoding: utf8
             path: classpath:db/changelog/202602051125.sql
   - changeSet:
+      id: 202602061650
+      author: DaGou12138
+      changes:
+        - sqlFile:
+            encoding: utf8
+            path: classpath:db/changelog/202602061650.sql
+  - changeSet:
+      id: 202602271137
+      author: RanChen
+      changes:
+        - sqlFile:
+            encoding: utf8
+            path: classpath:db/changelog/202602271137.sql
       id: 202602281000
       author: rainv123
       changes:

main/manager-api/src/main/resources/mapper/agent/AgentDao.xml

@@ -16,6 +16,10 @@
         <result column="llmModelId" property="llmModelId"/>
         <result column="ttsModelId" property="ttsModelId"/>
         <result column="ttsVoiceId" property="ttsVoiceId"/>
+        <result column="ttsLanguage" property="ttsLanguage"/>
+        <result column="ttsVolume" property="ttsVolume"/>
+        <result column="ttsRate" property="ttsRate"/>
+        <result column="ttsPitch" property="ttsPitch"/>
         <result column="memModelId" property="memModelId"/>
         <result column="intentModelId" property="intentModelId"/>
 
@@ -45,6 +49,10 @@
                a.vllm_model_id     AS vllmModelId,
                a.tts_model_id      AS ttsModelId,
                a.tts_voice_id      AS ttsVoiceId,
+               a.tts_language      AS ttsLanguage,
+               a.tts_volume        AS ttsVolume,
+               a.tts_rate          AS ttsRate,
+               a.tts_pitch         AS ttsPitch,
                a.mem_model_id      AS memModelId,
                a.intent_model_id   AS intentModelId,
                COALESCE(

main/manager-web/src/components/HeaderBar.vue

@@ -879,8 +879,8 @@ export default {
   }
 
   .equipment-management {
-    width: 79px;
+    min-width: 80px;
-    font-size: 9px;
+    font-size: 10px;
   }
 }
 

main/manager-web/src/components/TtsAdvancedSettings.vue

@@ -0,0 +1,240 @@
+<template>
+  <el-drawer
+    :visible.sync="drawerVisible"
+    :before-close="handleClose"
+    direction="rtl"
+    size="400px"
+    :modal="true"
+    :show-close="false"
+    custom-class="tts-advanced-drawer"
+  >
+    <div class="drawer-header" slot="title">
+      <span class="drawer-title">{{ $t('roleConfig.advancedSettings') }}</span>
+      <button class="drawer-close-btn" @click="handleClose">×</button>
+    </div>
+
+    <div class="drawer-content">
+      <el-form label-position="top">
+        <!-- 音量 -->
+        <el-form-item :label="$t('roleConfig.ttsVolume')">
+          <div class="slider-container">
+            <el-slider
+              v-model="localSettings.volume"
+              :min="-100"
+              :max="100"
+              :step="1"
+              :format-tooltip="formatTooltip"
+              class="tts-slider"
+            />
+            <span class="slider-hint">{{ $t('roleConfig.volumeHint') }}</span>
+          </div>
+        </el-form-item>
+
+        <!-- 语速 -->
+        <el-form-item :label="$t('roleConfig.ttsRate')">
+          <div class="slider-container">
+            <el-slider
+              v-model="localSettings.speed"
+              :min="-100"
+              :max="100"
+              :step="1"
+              :format-tooltip="formatTooltip"
+              class="tts-slider"
+            />
+            <span class="slider-hint">{{ $t('roleConfig.speedHint') }}</span>
+          </div>
+        </el-form-item>
+
+        <!-- 音调 -->
+        <el-form-item :label="$t('roleConfig.ttsPitch')">
+          <div class="slider-container">
+            <el-slider
+              v-model="localSettings.pitch"
+              :min="-100"
+              :max="100"
+              :step="1"
+              :format-tooltip="formatTooltip"
+              class="tts-slider"
+            />
+            <span class="slider-hint">{{ $t('roleConfig.pitchHint') }}</span>
+          </div>
+        </el-form-item>
+      </el-form>
+    </div>
+
+    <div class="drawer-footer">
+      <el-button @click="handleCancel">{{ $t('button.cancel') }}</el-button>
+      <el-button type="primary" @click="handleSave">{{ $t('button.save') }}</el-button>
+    </div>
+  </el-drawer>
+</template>
+
+<script>
+export default {
+  name: 'TtsAdvancedSettings',
+  props: {
+    visible: {
+      type: Boolean,
+      default: false
+    },
+    settings: {
+      type: Object,
+      default: () => ({
+        volume: 0,
+        speed: 0,
+        pitch: 0
+      })
+    }
+  },
+  data() {
+    return {
+      localSettings: {
+        volume: 0,
+        speed: 0,
+        pitch: 0
+      }
+    };
+  },
+  computed: {
+    drawerVisible: {
+      get() {
+        return this.visible;
+      },
+      set(val) {
+        this.$emit('update:visible', val);
+      }
+    }
+  },
+  watch: {
+    visible(newVal) {
+      if (newVal) {
+        // 当抽屉打开时，复制当前设置到本地
+        this.localSettings = { ...this.settings };
+      }
+    }
+  },
+  methods: {
+    handleClose() {
+      this.$emit('update:visible', false);
+    },
+    handleCancel() {
+      // 取消时不保存，直接关闭
+      this.handleClose();
+    },
+    handleSave() {
+      // 保存设置并关闭
+      this.$emit('save', { ...this.localSettings });
+      this.handleClose();
+    },
+    formatTooltip(val) {
+      return `${val}%`;
+    }
+  }
+};
+</script>
+
+<style scoped>
+.drawer-header {
+  display: flex;
+  align-items: center;
+  justify-content: space-between;
+  padding: 20px 24px;
+  border-bottom: 1px solid #e8f0ff;
+}
+
+.drawer-title {
+  font-size: 18px;
+  font-weight: 600;
+  color: #3d4566;
+}
+
+.drawer-close-btn {
+  width: 32px;
+  height: 32px;
+  border-radius: 50%;
+  border: 2px solid #cfcfcf;
+  background: none;
+  font-size: 28px;
+  font-weight: lighter;
+  color: #cfcfcf;
+  cursor: pointer;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  padding: 0;
+  outline: none;
+  transition: all 0.3s;
+}
+
+.drawer-close-btn:hover {
+  color: #409eff;
+  border-color: #409eff;
+}
+
+.drawer-content {
+  padding: 24px;
+  flex: 1;
+  overflow-y: auto;
+}
+
+.slider-container {
+  width: 100%;
+}
+
+.slider-hint {
+  display: block;
+  font-size: 12px;
+  color: #909399;
+  margin-top: 8px;
+  line-height: 1.5;
+}
+
+.tts-slider {
+  width: 100%;
+}
+
+.tts-slider ::v-deep .el-slider__input {
+  width: 80px;
+}
+
+.tts-slider ::v-deep .el-input__inner {
+  text-align: center;
+  padding: 0 8px;
+}
+
+.drawer-footer {
+  padding: 16px 24px;
+  border-top: 1px solid #e8f0ff;
+  display: flex;
+  justify-content: center;
+  gap: 12px;
+}
+
+.drawer-footer .el-button {
+  min-width: 80px;
+}
+
+::v-deep .el-form-item__label {
+  font-size: 14px !important;
+  color: #3d4566 !important;
+  font-weight: 500;
+  padding-bottom: 8px;
+}
+
+::v-deep .el-form-item {
+  margin-bottom: 24px;
+}
+</style>
+
+<style>
+.tts-advanced-drawer .el-drawer__header {
+  margin-bottom: 0;
+  padding: 0;
+}
+
+.tts-advanced-drawer .el-drawer__body {
+  display: flex;
+  flex-direction: column;
+  padding: 0;
+}
+</style>

main/manager-web/src/i18n/de.js

@@ -764,7 +764,16 @@ export default {
   'roleConfig.memoryHis': 'Speicher',
   'roleConfig.memory': 'Speicher',
   'roleConfig.intent': 'Intent',
+  'roleConfig.language': 'Sprache auswählen',
   'roleConfig.voiceType': 'Stimmtyp',
+  'roleConfig.ttsVolume': 'Lautstärke',
+  'roleConfig.ttsRate': 'Geschwindigkeit',
+  'roleConfig.ttsPitch': 'Tonhöhe',
+  'roleConfig.ttsAdvanced': 'TTS-Parameter',
+  'roleConfig.advancedSettings': 'Erweiterte Einstellungen',
+  'roleConfig.volumeHint': '-100=Min, 0=Standard, 100=Max',
+  'roleConfig.speedHint': '-100=Langsamste, 0=Standard, 100=Schnellste',
+  'roleConfig.pitchHint': '-100=Niedrigste, 0=Standard, 100=Höchste',
   'roleConfig.pleaseEnterContent': 'Bitte Inhalt eingeben',
   'roleConfig.pleaseEnterLangCode': 'Bitte Sprachcode eingeben, z.B.: en_US',
   'roleConfig.pleaseEnterLangName': 'Bitte Interaktionssprache eingeben, z.B.: Englisch',

main/manager-web/src/i18n/en.js

@@ -764,7 +764,16 @@ export default {
   'roleConfig.memoryHis': 'Memory',
   'roleConfig.memory': 'Memory Model',
   'roleConfig.intent': 'Intent Recognition',
+  'roleConfig.language': 'Select Language',
   'roleConfig.voiceType': 'Voice Type',
+  'roleConfig.ttsVolume': 'Volume',
+  'roleConfig.ttsRate': 'Speed',
+  'roleConfig.ttsPitch': 'Pitch',
+  'roleConfig.ttsAdvanced': 'TTS Parameters',
+  'roleConfig.advancedSettings': 'Advanced Settings',
+  'roleConfig.volumeHint': '-100=Min, 0=Standard, 100=Max',
+  'roleConfig.speedHint': '-100=Slowest, 0=Standard, 100=Fastest',
+  'roleConfig.pitchHint': '-100=Lowest, 0=Standard, 100=Highest',
   'roleConfig.pleaseEnterContent': 'Please enter content',
   'roleConfig.pleaseEnterLangCode': 'Please enter language code, e.g.: en_US',
   'roleConfig.pleaseEnterLangName': 'Please enter interaction language, e.g.: English',

main/manager-web/src/i18n/vi.js

@@ -764,7 +764,16 @@ export default {
   'roleConfig.memoryHis': 'Bộ nhớ',
   'roleConfig.memory': 'Mô hình bộ nhớ',
   'roleConfig.intent': 'Nhận dạng ý định',
+  'roleConfig.language': 'Chọn ngôn ngữ',
   'roleConfig.voiceType': 'Loại giọng nói',
+  'roleConfig.ttsVolume': 'Âm lượng',
+  'roleConfig.ttsRate': 'Tốc độ',
+  'roleConfig.ttsPitch': 'Cao độ',
+  'roleConfig.ttsAdvanced': 'Tham số TTS',
+  'roleConfig.advancedSettings': 'Cài đặt nâng cao',
+  'roleConfig.volumeHint': '-100=Tối thiểu, 0=Tiêu chuẩn, 100=Tối đa',
+  'roleConfig.speedHint': '-100=Chậm nhất, 0=Tiêu chuẩn, 100=Nhanh nhất',
+  'roleConfig.pitchHint': '-100=Thấp nhất, 0=Tiêu chuẩn, 100=Cao nhất',
   'roleConfig.pleaseEnterContent': 'Vui lòng nhập nội dung',
   'roleConfig.pleaseEnterLangCode': 'Vui lòng nhập mã ngôn ngữ, ví dụ: en_US',
   'roleConfig.pleaseEnterLangName': 'Vui lòng nhập ngôn ngữ tương tác, ví dụ: Tiếng Anh',

main/manager-web/src/i18n/zh_CN.js

@@ -764,7 +764,16 @@ export default {
   'roleConfig.memoryHis': '记忆',
   'roleConfig.memory': '记忆模式',
   'roleConfig.tts': '语音合成(TTS)',
+  'roleConfig.language': '选择语言',
   'roleConfig.voiceType': '声音音色(Voice)',
+  'roleConfig.ttsVolume': '音量',
+  'roleConfig.ttsRate': '语速',
+  'roleConfig.ttsPitch': '音调',
+  'roleConfig.ttsAdvanced': 'TTS参数',
+  'roleConfig.advancedSettings': '高级设置',
+  'roleConfig.volumeHint': '-100=最小, 0=标准, 100=最大',
+  'roleConfig.speedHint': '-100=最慢, 0=标准, 100=最快',
+  'roleConfig.pitchHint': '-100=最低, 0=标准, 100=最高',
   'roleConfig.pleaseEnterContent': '请输入内容',
   'roleConfig.pleaseEnterLangCode': '请输入语言编码，如：zh_CN',
   'roleConfig.pleaseEnterLangName': '请输入交互语种，如：中文',

main/manager-web/src/i18n/zh_TW.js

@@ -764,7 +764,16 @@ export default {
   'roleConfig.memoryHis': '記憶',
   'roleConfig.memory': '記憶模式',
   'roleConfig.intent': '意圖識別(Intent)',
+  'roleConfig.language': '選擇語言',
   'roleConfig.voiceType': '聲音音色(Voice)',
+  'roleConfig.ttsVolume': '音量',
+  'roleConfig.ttsRate': '語速',
+  'roleConfig.ttsPitch': '音調',
+  'roleConfig.ttsAdvanced': 'TTS參數',
+  'roleConfig.advancedSettings': '高級設置',
+  'roleConfig.volumeHint': '-100=最小, 0=標準, 100=最大',
+  'roleConfig.speedHint': '-100=最慢, 0=標準, 100=最快',
+  'roleConfig.pitchHint': '-100=最低, 0=標準, 100=最高',
   'roleConfig.pleaseEnterContent': '請輸入內容',
   'roleConfig.pleaseEnterLangCode': '請輸入語言編碼，如：zh_TW',
   'roleConfig.pleaseEnterLangName': '請輸入交互語種，如：繁體中文',

main/manager-web/src/views/roleConfig.vue

@@ -256,47 +256,78 @@
                         </div>
                       </div>
                     </el-form-item>
-                    <el-form-item :label="$t('roleConfig.voiceType')">
+                    <div class="model-row">
-                      <el-select
+                      <!-- 语言筛选器 -->
-                        v-model="form.ttsVoiceId"
+                      <el-form-item :label="$t('roleConfig.language')" class="model-item language-select-item">
-                        filterable
+                        <div class="model-select-wrapper">
-                        :placeholder="$t('roleConfig.pleaseSelect')"
+                          <el-select
-                        class="form-select"
+                            v-model="selectedLanguage"
-                      >
+                            :placeholder="$t('roleConfig.selectLanguage')"
-                        <el-option
+                            class="form-select language-select"
-                          v-for="(item, index) in voiceOptions"
+                            @change="filterVoicesByLanguage"
-                          :key="`voice-${index}`"
-                          :label="item.label"
-                          :value="item.value"
-                        >
-                          <div
-                            style="
-                              display: flex;
-                              justify-content: space-between;
-                              align-items: center;
-                            "
                           >
-                            <span>{{ item.label }}</span>
+                            <el-option
-                            <template v-if="hasAudioPreview(item)">
+                              v-for="(lang, index) in languageOptions"
-                              <el-button
+                              :key="`lang-${index}`"
-                                type="text"
+                              :label="lang.label"
-                                :icon="
+                              :value="lang.value"
-                                  playingVoice &&
+                            />
-                                  currentPlayingVoiceId === item.value &&
+                          </el-select>
-                                  !isPaused
+                        </div>
-                                    ? 'el-icon-video-pause'
+                      </el-form-item>
-                                    : 'el-icon-video-play'
+
+                      <!-- 音色选择器 -->
+                      <el-form-item :label="$t('roleConfig.voiceType')" class="model-item">
+                        <div class="model-select-wrapper">
+                          <el-select
+                            v-model="form.ttsVoiceId"
+                            filterable
+                            :placeholder="$t('roleConfig.pleaseSelect')"
+                            class="form-select"
+                          >
+                            <el-option
+                              v-for="(item, index) in voiceOptions"
+                              :key="`voice-${index}`"
+                              :label="item.label"
+                              :value="item.value"
+                            >
+                              <div
+                                style="
+                                  display: flex;
+                                  justify-content: space-between;
+                                  align-items: center;
                                 "
-                                size="small"
+                              >
-                                @click.stop="toggleAudioPlayback(item.value)"
+                                <span>{{ item.label }}</span>
-                                :loading="false"
+                                <template v-if="hasAudioPreview(item)">
-                                class="play-button"
+                                  <el-button
-                              />
+                                    type="text"
-                            </template>
+                                    :icon="
-                          </div>
+                                      playingVoice &&
-                        </el-option>
+                                      currentPlayingVoiceId === item.value &&
-                      </el-select>
+                                      !isPaused
-                    </el-form-item>
+                                        ? 'el-icon-video-pause'
+                                        : 'el-icon-video-play'
+                                    "
+                                    size="small"
+                                    @click.stop="toggleAudioPlayback(item.value)"
+                                    :loading="false"
+                                    class="play-button"
+                                  />
+                                </template>
+                              </div>
+                            </el-option>
+                          </el-select>
+                          <el-button
+                            class="edit-function-btn"
+                            style="margin-left: 10px;"
+                            @click="openTtsAdvancedSettings"
+                          >
+                            {{ $t('roleConfig.advancedSettings') }}
+                          </el-button>
+                        </div>
+                      </el-form-item>
+                    </div>
                   </div>
                 </div>
               </div>
@@ -318,6 +349,11 @@
       :providers="currentContextProviders"
       @confirm="handleUpdateContext"
     />
+    <tts-advanced-settings
+      :visible.sync="showTtsAdvancedDialog"
+      :settings="ttsSettings"
+      @save="handleTtsSettingsSave"
+    />
   </div>
 </template>
 
@@ -327,20 +363,30 @@ import { getServiceUrl } from "@/apis/api";
 import RequestService from "@/apis/httpRequest";
 import FunctionDialog from "@/components/FunctionDialog.vue";
 import ContextProviderDialog from "@/components/ContextProviderDialog.vue";
+import TtsAdvancedSettings from "@/components/TtsAdvancedSettings.vue";
 import HeaderBar from "@/components/HeaderBar.vue";
 import i18n from "@/i18n";
 import featureManager from "@/utils/featureManager"; 
 
 export default {
   name: "RoleConfigPage",
-  components: { HeaderBar, FunctionDialog, ContextProviderDialog },
+  components: { HeaderBar, FunctionDialog, ContextProviderDialog, TtsAdvancedSettings },
   data() {
     return {
       showContextProviderDialog: false,
+      showTtsAdvancedDialog: false,
+      ttsSettings: {
+        volume: 0,
+        speed: 0,
+        pitch: 0
+      },
       form: {
         agentCode: "",
         agentName: "",
         ttsVoiceId: "",
+        ttsVolume: null,
+        ttsRate: null,
+        ttsPitch: null,
         chatHistoryConf: 0,
         systemPrompt: "",
         summaryMemory: "",
@@ -381,6 +427,9 @@ export default {
       isPaused: false,
       currentAudio: null,
       currentPlayingVoiceId: null,
+      // 语言筛选相关状态
+      languageOptions: [], // 语言选项列表
+      selectedLanguage: '', // 当前选中的语言
       // 功能状态
       featureStatus: {
         vad: false, // 语言检测活动功能状态
@@ -412,6 +461,7 @@ export default {
         vllmModelId: this.form.model.vllmModelId,
         ttsModelId: this.form.model.ttsModelId,
         ttsVoiceId: this.form.ttsVoiceId,
+        ttsLanguage: this.selectedLanguage,
         chatHistoryConf: this.form.chatHistoryConf,
         memModelId: this.form.model.memModelId,
         intentModelId: this.form.model.intentModelId,
@@ -428,6 +478,17 @@ export default {
         }),
         contextProviders: this.currentContextProviders,
       };
+
+      // 只在用户设置了TTS参数时才传递（不为null/undefined）
+      if (this.form.ttsVolume !== null && this.form.ttsVolume !== undefined) {
+        configData.ttsVolume = this.form.ttsVolume;
+      }
+      if (this.form.ttsRate !== null && this.form.ttsRate !== undefined) {
+        configData.ttsRate = this.form.ttsRate;
+      }
+      if (this.form.ttsPitch !== null && this.form.ttsPitch !== undefined) {
+        configData.ttsPitch = this.form.ttsPitch;
+      }
       Api.agent.updateAgentConfig(this.$route.query.agentId, configData, ({ data }) => {
         if (data.code === 0) {
           this.$message.success({
@@ -543,6 +604,14 @@ export default {
               intentModelId: data.data.intentModelId,
             },
           };
+
+          // 同步TTS设置到ttsSettings
+          this.ttsSettings = {
+            volume: this.form.ttsVolume || 0,
+            speed: this.form.ttsRate || 0,
+            pitch: this.form.ttsPitch || 0
+          };
+
           // 后端只给了最小映射：[{ id, agentId, pluginId }, ...]
           const savedMappings = data.data.functions || [];
           
@@ -628,32 +697,91 @@ export default {
       if (!modelId) {
         this.voiceOptions = [];
         this.voiceDetails = {};
+        this.languageOptions = [];
+        this.selectedLanguage = '';
         return;
       }
       Api.model.getModelVoices(modelId, "", ({ data }) => {
         if (data.code === 0 && data.data) {
-          this.voiceOptions = data.data.map((voice) => ({
+          // 保存完整的音色信息
-            value: voice.id,
-            label: voice.name,
-            // 只保留后端实际返回的音频相关字段
-            voiceDemo: voice.voiceDemo,
-            voice_demo: voice.voice_demo,
-            // 使用后端实际返回的 isClone 字段
-            isClone: Boolean(voice.isClone),
-            // 保存训练状态字段
-            train_status: voice.trainStatus,
-          }));
-          // 保存完整的音色信息，添加调试信息
           this.voiceDetails = data.data.reduce((acc, voice) => {
             acc[voice.id] = voice;
             return acc;
           }, {});
+          
+          // 提取所有语言选项并去重
+          const allLanguages = new Set();
+          data.data.forEach(voice => {
+            if (voice.languages) {
+              const languagesArray = voice.languages.split('、').map(lang => lang.trim()).filter(lang => lang);
+              languagesArray.forEach(lang => allLanguages.add(lang));
+            }
+          });
+
+          this.languageOptions = Array.from(allLanguages).map(lang => ({
+            value: lang,
+            label: lang
+          }));
+
+          // 使用后端返回的用户选择的语言，如果没有则使用第一个语言选项
+          if (this.form.ttsLanguage && this.languageOptions.some(option => option.value === this.form.ttsLanguage)) {
+            this.selectedLanguage = this.form.ttsLanguage;
+          } else if (this.languageOptions.length > 0) {
+            this.selectedLanguage = this.languageOptions[0].value;
+          }
+
+          // 根据选中的语言筛选音色
+          this.filterVoicesByLanguage();
         } else {
           this.voiceOptions = [];
           this.voiceDetails = {};
+          this.languageOptions = [];
+          this.selectedLanguage = '';
         }
       });
     },
+    
+    // 根据语言筛选音色
+    filterVoicesByLanguage() {
+      if (!this.voiceDetails || Object.keys(this.voiceDetails).length === 0) {
+        this.voiceOptions = [];
+        return;
+      }
+
+      const allVoices = Object.values(this.voiceDetails);
+
+      // 根据选中的语言筛选音色
+      const filteredVoices = allVoices.filter(voice => {
+        if (!voice.languages) return false;
+        const languagesArray = voice.languages.split('、').map(lang => lang.trim()).filter(lang => lang);
+        return languagesArray.includes(this.selectedLanguage);
+      });
+
+      this.voiceOptions = filteredVoices.map((voice) => ({
+        value: voice.id,
+        label: voice.name,
+        voiceDemo: voice.voiceDemo,
+        voice_demo: voice.voice_demo,
+        isClone: Boolean(voice.isClone),
+        train_status: voice.trainStatus,
+      }));
+
+      // 检查当前选中的音色是否支持当前语言，如果不支持则选择第一个
+      const currentVoiceSupportsLanguage = this.form.ttsVoiceId &&
+        filteredVoices.some(voice => voice.id === this.form.ttsVoiceId);
+
+      if (!currentVoiceSupportsLanguage) {
+        this.form.ttsVoiceId = filteredVoices.length > 0 ? filteredVoices[0].id : '';
+      }
+
+      // 同步到ttsSettings（如果值为null，使用0作为显示默认值，但不修改form中的值）
+      this.ttsSettings = {
+        volume: this.form.ttsVolume !== null && this.form.ttsVolume !== undefined ? this.form.ttsVolume : 0,
+        speed: this.form.ttsRate !== null && this.form.ttsRate !== undefined ? this.form.ttsRate : 0,
+        pitch: this.form.ttsPitch !== null && this.form.ttsPitch !== undefined ? this.form.ttsPitch : 0
+      };
+    },
+
     getFunctionDisplayChar(name) {
       if (!name || name.length === 0) return "";
 
@@ -719,6 +847,16 @@ export default {
     openContextProviderDialog() {
       this.showContextProviderDialog = true;
     },
+    openTtsAdvancedSettings() {
+      this.showTtsAdvancedDialog = true;
+    },
+    handleTtsSettingsSave(settings) {
+      // 保存TTS设置
+      this.ttsSettings = { ...settings };
+      this.form.ttsVolume = settings.volume;
+      this.form.ttsRate = settings.speed;
+      this.form.ttsPitch = settings.pitch;
+    },
     handleUpdateContext(providers) {
       this.currentContextProviders = providers;
     },
@@ -1354,6 +1492,15 @@ export default {
   margin-bottom: 0;
 }
 
+.model-row .language-select-item {
+  flex: 0 0 35%;
+  max-width: 35%;
+}
+
+.model-row .language-select-item .language-select {
+  width: 100%;
+}
+
 .model-row .el-form-item__label {
   font-size: 12px !important;
   color: #3d4566 !important;
@@ -1529,6 +1676,31 @@ export default {
     text-decoration: underline;
   }
 }
+
+.slider-wrapper {
+  width: 100%;
+  padding-right: 12px;
+}
+
+.slider-hint {
+  display: block;
+  font-size: 12px;
+  color: #909399;
+  margin-top: 4px;
+  line-height: 1.5;
+}
+
+.tts-slider {
+  width: 100%;
+}
+
+.tts-slider ::v-deep .el-slider__input {
+  width: 80px;
+}
+
+.tts-slider ::v-deep .el-input__inner {
+  text-align: center;
+  padding: 0 8px;
 .input-new-tag {
   width: 90px;
   &::v-deep(.el-input__inner) {

main/xiaozhi-server/agent-base-prompt.txt

@@ -2,6 +2,10 @@
 {{base_prompt}}
 </identity>
 
+<language>
+【语言规范】你必须使用{{language}}进行回复和交流。无论用户使用何种语言提问，你都应该用{{language}}来回答。
+</language>
+
 <emotion>
 【核心目标】你不是冰冷的机器！请敏锐感知用户情绪，作为懂人心的伙伴，用有温度的回应照亮对话。
 - **情感融入：**

main/xiaozhi-server/config.yaml

@@ -733,6 +733,7 @@ TTS:
     type: edge
     voice: zh-CN-XiaoxiaoNeural
     output_dir: tmp/
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   DoubaoTTS:
     # 定义TTS API类型
     type: doubao
@@ -751,6 +752,7 @@ TTS:
     speed_ratio: 1.0
     volume_ratio: 1.0
     pitch_ratio: 1.0
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   #火山tts，支持双向流式tts
   HuoshanDoubleStreamTTS:
     type: huoshan_double_stream
@@ -791,6 +793,7 @@ TTS:
     #       mix_factor: 0.3
     #     - source_speaker: zh_male_ahu_conversation_wvae_bigtts
     #       mix_factor: 0.4
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   CosyVoiceSiliconflow:
     type: siliconflow
     # 硅基流动TTS
@@ -800,6 +803,7 @@ TTS:
     output_dir: tmp/
     access_token: 你的硅基流动API密钥
     response_format: wav
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   CozeCnTTS:
     type: cozecn
     # COZECN TTS
@@ -808,6 +812,10 @@ TTS:
     output_dir: tmp/
     access_token: 你的coze web key
     response_format: wav
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
+    # 以下可不用设置，使用默认设置
+    # speed: 1  # 语速：-0.5到2
+    # loudness_rate: 0  # 音量：-50到100
   VolcesAiGatewayTTS:
     type: openai
     # 火山引擎 - 边缘大模型网关
@@ -822,6 +830,7 @@ TTS:
     voice: zh_male_shaonianzixin_moon_bigtts
     speed: 1
     output_dir: tmp/
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   FishSpeech:
     # 参照教程：https://github.com/xinnan-tech/xiaozhi-esp32-server/blob/main/docs/fish-speech-integration.md
     type: fishspeech
@@ -843,6 +852,7 @@ TTS:
     rate: 44100
     api_key: "你的api_key"
     api_url: "http://127.0.0.1:8080/v1/tts"
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   GPT_SOVITS_V2:
     # 定义TTS API类型
     #启动tts方法：
@@ -868,6 +878,7 @@ TTS:
     parallel_infer: true
     repetition_penalty: 1.35
     aux_ref_audio_paths: []
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   GPT_SOVITS_V3:
     # 定义TTS API类型 GPT-SoVITS-v3lora-20250228
     #启动tts方法：
@@ -887,6 +898,7 @@ TTS:
     inp_refs: []
     sample_steps: 32
     if_sr: false
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   MinimaxTTSHTTPStream:
   # Minimax流式语音合成服务
     type: minimax_httpstream
@@ -918,6 +930,7 @@ TTS:
     #     voice_id: female-shaonv
     #     weight: 1
     # language_boost: auto
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   AliyunTTS:
     # 阿里云智能语音交互服务，需要先在阿里云平台开通服务，然后获取验证信息
     # 平台地址：https://nls-portal.console.aliyun.com/
@@ -937,6 +950,7 @@ TTS:
     # volume: 50
     # speech_rate: 0
     # pitch_rate: 0
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   AliyunStreamTTS:
     # 阿里云CosyVoice大模型流式文本语音合成
     # 采用FlowingSpeechSynthesizer接口，支持更低延迟和更自然的语音质量
@@ -960,6 +974,7 @@ TTS:
     # volume: 50  # 音量：0-100
     # speech_rate: 0  # 语速：-500到500
     # pitch_rate: 0  # 语调：-500到500
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   TencentTTS:
     # 腾讯云智能语音交互服务，需要先在腾讯云平台开通服务
     # appid、secret_id、secret_key申请地址：https://console.cloud.tencent.com/cam/capi
@@ -971,7 +986,11 @@ TTS:
     secret_key: 你的腾讯云SecretKey
     region: ap-guangzhou
     voice: 101001
-
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
+    # 以下可不用设置，使用默认设置
+    # format: wav  # 音频格式：pcm、wav、mp3
+    # volume: 0  # 音量：-10到10
+    # speech_rate: 0  # 语速：-2到6
   TTS302AI:
     # 302AI语音合成服务，需要先在302平台创建账户充值，并获取密钥信息
     # 添加 302.ai TTS 配置
@@ -985,6 +1004,7 @@ TTS:
     voice: "zh_female_wanwanxiaohe_moon_bigtts"
     output_dir: tmp/
     access_token: "你的302API密钥"
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   GizwitsTTS:
     type: doubao
     # 火山引擎作为基座，可以完全使用企业级火山引擎语音合成服务
@@ -996,6 +1016,7 @@ TTS:
     voice: "zh_female_wanwanxiaohe_moon_bigtts"
     output_dir: tmp/
     access_token: "你的机智云API key"
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   ACGNTTS:
     #在线网址：https://acgn.ttson.cn/
     #token购买：www.ttson.cn
@@ -1013,6 +1034,7 @@ TTS:
     format: mp3
     output_dir: tmp/
     emotion: 1
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   OpenAITTS:
     # openai官方文本转语音服务，可支持全球大多数语种
     type: openai
@@ -1028,6 +1050,7 @@ TTS:
     # 语速范围0.25-4.0
     speed: 1
     output_dir: tmp/
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   CustomTTS:
     # 自定义的TTS接口服务，请求参数可自定义，可接入众多TTS服务
     # 以本地部署的KokoroTTS为例
@@ -1050,6 +1073,7 @@ TTS:
       # Authorization: Bearer xxxx
     format: mp3 # 接口返回的音频格式
     output_dir: tmp/
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   LinkeraiTTS:
     type: linkerai
     api_url: https://tts.linkerai.cn/tts
@@ -1061,6 +1085,7 @@ TTS:
     access_token: "U4YdYXVfpwWnk2t5Gp822zWPCuORyeJL"
     voice: "OUeAo1mhq6IBExi"
     output_dir: tmp/
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   PaddleSpeechTTS:
     #百度飞浆 PaddleSpeech 支持本地离线部署 支持模型训练
     #框架地址 https://www.paddlepaddle.org.cn/
@@ -1074,6 +1099,7 @@ TTS:
     speed: 1.0  # 语速，1.0 表示正常语速，>1 表示加快，<1 表示减慢
     volume: 1.0  # 音量，1.0 表示正常音量，>1 表示增大，<1 表示减小
     save_path:   # 保存路径
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   IndexStreamTTS:
     # 基于Index-TTS-vLLM项目的TTS接口服务
     # 参照教程：https://github.com/Ksuriuri/index-tts-vllm/blob/master/README.md
@@ -1083,6 +1109,7 @@ TTS:
     # 默认音色，如需其他音色可到项目assets文件夹下注册
     voice: "jay_klee"
     output_dir: tmp/
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   AliBLTTS:
     # 阿里百炼CosyVoice大模型流式文本语音合成
     # 可在这里找到你的 api_key https://bailian.console.aliyun.com/?apiKey=1#/api-key
@@ -1097,6 +1124,7 @@ TTS:
     # volume: 50  # 音量：0-100
     # rate: 1  # 语速：0.5~2
     # pitch: 1  # 语调：0.5~2
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文
   XunFeiTTS:
     # 讯飞TTS服务 官方网站：https://www.xfyun.cn/
     # 登录讯飞语音技术平台 https://console.xfyun.cn/app/myapp 创建相关应用
@@ -1118,3 +1146,4 @@ TTS:
     # volume: 50  # 音量：0-100
     # speed: 50  # 语速：0-100
     # pitch: 50  # 语调：0-100
+    # language: "中文"  # 指定输出语种,如:中文、英语、日语、韩语等,请根据所选音色支持的语言进行设置,不填则默认为中文

main/xiaozhi-server/core/connection.py

@@ -134,7 +134,6 @@ class ConnectionHandler:
         self.asr_audio = []
         self.asr_audio_queue = queue.Queue()
         self.current_speaker = None  # 存储当前说话人
-        self.current_language_tag = None  # 存储当前ASR识别的语言标签
 
         # llm相关变量
         self.dialogue = Dialogue()

main/xiaozhi-server/core/handle/receiveAudioHandle.py

@@ -67,11 +67,6 @@ async def startToChat(conn: "ConnectionHandler", text):
         conn.current_speaker = speaker_name
     else:
         conn.current_speaker = None
-    # 保存语种信息到连接对象
-    if language_tag:
-        conn.current_language_tag = language_tag
-    else:
-        conn.current_language_tag = "zh"
 
     if conn.need_bind:
         await check_bind_device(conn)

main/xiaozhi-server/core/providers/tts/alibl_stream.py

@@ -6,10 +6,10 @@ import queue
 import asyncio
 import traceback
 import websockets
-from typing import Callable, Any
+
 from asyncio import Task
+from typing import Callable, Any
 from config.logger import setup_logging
-from core.utils import opus_encoder_utils
 from core.utils.tts import MarkdownCleaner
 from core.providers.tts.base import TTSProviderBase
 from core.providers.tts.dto.dto import SentenceType, ContentType, InterfaceType
@@ -19,6 +19,12 @@ logger = setup_logging()
 
 
 class TTSProvider(TTSProviderBase):
+    TTS_PARAM_CONFIG = [
+        ("ttsVolume", "volume", 0, 100, 50, int),
+        ("ttsRate", "rate", 0.5, 2.0, 1.0, lambda v: round(v, 1)),
+        ("ttsPitch", "pitch", 0.5, 2.0, 1.0, lambda v: round(v, 1)),
+    ]
+
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
 
@@ -52,6 +58,9 @@ class TTSProvider(TTSProviderBase):
         pitch = config.get("pitch", "1.0")
         self.pitch = float(pitch) if pitch else 1.0
 
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        self._apply_percentage_params(config)
+
         self.header = {
             "Authorization": f"Bearer {self.api_key}",
             # "user-agent": "your_platform_info", // 可选

main/xiaozhi-server/core/providers/tts/aliyun.py

@@ -1,15 +1,17 @@
 import uuid
 import json
 import hmac
+import time
 import hashlib
 import base64
 import requests
-from datetime import datetime
+
-from core.providers.tts.base import TTSProviderBase
-from config.logger import setup_logging
-import time
-import uuid
 from urllib import parse
+from datetime import datetime
+from config.logger import setup_logging
+from core.providers.tts.base import TTSProviderBase
+from core.utils.tts import convert_percentage_to_range
+
 
 TAG = __name__
 logger = setup_logging()
@@ -84,6 +86,11 @@ class AccessToken:
 
 
 class TTSProvider(TTSProviderBase):
+    TTS_PARAM_CONFIG = [
+        ("ttsVolume", "volume", 0, 100, 50, int),
+        ("ttsRate", "speech_rate", -500, 500, 0, int),
+        ("ttsPitch", "pitch_rate", -500, 500, 0, int),
+    ]
 
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
@@ -93,7 +100,6 @@ class TTSProvider(TTSProviderBase):
         self.access_key_secret = config.get("access_key_secret")
 
         self.appkey = config.get("appkey")
-        self.format = config.get("format", "wav")
         self.audio_file_type = config.get("format", "wav")
 
         if config.get("private_voice"):
@@ -110,6 +116,9 @@ class TTSProvider(TTSProviderBase):
         pitch_rate = config.get("pitch_rate", "0")
         self.pitch_rate = int(pitch_rate) if pitch_rate else 0
 
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        self._apply_percentage_params(config)
+
         self.host = config.get("host", "nls-gateway-cn-shanghai.aliyuncs.com")
         self.api_url = f"https://{self.host}/stream/v1/tts"
         self.header = {"Content-Type": "application/json"}
@@ -169,7 +178,7 @@ class TTSProvider(TTSProviderBase):
             "appkey": self.appkey,
             "token": self.token,
             "text": text,
-            "format": self.format,
+            "format": self.audio_file_type,
             "sample_rate": self.conn.sample_rate,
             "voice": self.voice,
             "volume": self.volume,

main/xiaozhi-server/core/providers/tts/aliyun_stream.py

@@ -1,4 +1,4 @@
-import random
+import os
 import uuid
 import json
 import hmac
@@ -8,17 +8,17 @@ import time
 import queue
 import asyncio
 import traceback
-from typing import Callable, Any
-from asyncio import Task
 import websockets
-import os
+
-from datetime import datetime
+from asyncio import Task
 from urllib import parse
+from datetime import datetime
+from typing import Callable, Any
+from config.logger import setup_logging
+from core.utils.tts import MarkdownCleaner
 from core.providers.tts.base import TTSProviderBase
 from core.providers.tts.dto.dto import SentenceType, ContentType, InterfaceType
-from core.utils.tts import MarkdownCleaner
+
-from core.utils import opus_encoder_utils, textUtils
-from config.logger import setup_logging
 
 TAG = __name__
 logger = setup_logging()
@@ -87,6 +87,12 @@ class AccessToken:
 
 
 class TTSProvider(TTSProviderBase):
+    TTS_PARAM_CONFIG = [
+        ("ttsVolume", "volume", 0, 100, 50, int),
+        ("ttsRate", "speech_rate", -500, 500, 0, int),
+        ("ttsPitch", "pitch_rate", -500, 500, 0, int),
+    ]
+
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
 
@@ -98,7 +104,6 @@ class TTSProvider(TTSProviderBase):
         self.access_key_secret = config.get("access_key_secret")
         self.appkey = config.get("appkey")
         self.format = config.get("format", "pcm")
-        self.audio_file_type = config.get("format", "pcm")
 
         # 音色配置 - CosyVoice大模型音色
         if config.get("private_voice"):
@@ -116,6 +121,9 @@ class TTSProvider(TTSProviderBase):
         pitch_rate = config.get("pitch_rate", "0")
         self.pitch_rate = int(pitch_rate) if pitch_rate else 0
 
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        self._apply_percentage_params(config)
+
         # WebSocket配置
         self.host = config.get("host", "nls-gateway-cn-beijing.aliyuncs.com")
         # 如果配置的是内网地址（包含-internal.aliyuncs.com），则使用ws协议，默认是wss协议

main/xiaozhi-server/core/providers/tts/base.py

@@ -13,7 +13,7 @@ from typing import Callable, Any
 from abc import ABC, abstractmethod
 from config.logger import setup_logging
 from core.utils import opus_encoder_utils
-from core.utils.tts import MarkdownCleaner
+from core.utils.tts import MarkdownCleaner, convert_percentage_to_range
 from core.utils.output_counter import add_device_output
 from core.handle.reportHandle import enqueue_tts_report
 from core.handle.sendAudioHandle import sendAudioMessage
@@ -463,3 +463,10 @@ class TTSProviderBase(ABC):
                 self.processed_chars += len(full_text)
                 return True
         return False
+
+    def _apply_percentage_params(self, config):
+        """根据子类定义的 TTS_PARAM_CONFIG 批量应用百分比参数"""
+        for config_key, attr_name, min_val, max_val, base_val, transform in self.TTS_PARAM_CONFIG:
+            if config_key in config:
+                val = convert_percentage_to_range(config[config_key], min_val, max_val, base_val)
+                setattr(self, attr_name, transform(val) if transform else val)

main/xiaozhi-server/core/providers/tts/cozecn.py

@@ -3,6 +3,11 @@ from core.providers.tts.base import TTSProviderBase
 
 
 class TTSProvider(TTSProviderBase):
+    TTS_PARAM_CONFIG = [
+        ("speed", "speed", 0.5, 2, 1, lambda v: round(float(v), 1)),
+        ("loudness_rate", "loudness_rate", -50, 100, 0, int),
+    ]
+
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
         self.model = config.get("model")
@@ -11,17 +16,29 @@ class TTSProvider(TTSProviderBase):
             self.voice = config.get("private_voice")
         else:
             self.voice = config.get("voice")
-        self.response_format = config.get("response_format", "wav")
         self.audio_file_type = config.get("response_format", "wav")
         self.host = "api.coze.cn"
         self.api_url = f"https://{self.host}/v1/audio/speech"
 
+        # 音频参数配置
+        speed = config.get("speed", "0")
+        self.speed = int(speed) if speed else 0
+
+        loudness_rate = config.get("loudness_rate", "0")
+        self.loudness_rate = int(loudness_rate) if loudness_rate else 0
+
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        self._apply_percentage_params(config)
+
     async def text_to_speak(self, text, output_file):
         request_json = {
             "model": self.model,
             "input": text,
             "voice_id": self.voice,
-            "response_format": self.response_format,
+            "response_format": self.audio_file_type,
+            "sample_rate": self.conn.sample_rate,
+            "speed": self.speed,
+            "loudness_rate": self.loudness_rate,
         }
         headers = {
             "Authorization": f"Bearer {self.access_token}",

main/xiaozhi-server/core/providers/tts/custom.py

@@ -15,7 +15,6 @@ class TTSProvider(TTSProviderBase):
         self.url = config.get("url")
         self.method = config.get("method", "GET")
         self.headers = config.get("headers", {})
-        self.format = config.get("format", "wav")
         self.audio_file_type = config.get("format", "wav")
         self.output_file = config.get("output_dir", "tmp/")
         self.params = config.get("params")
@@ -29,7 +28,7 @@ class TTSProvider(TTSProviderBase):
             raise TypeError("Custom TTS配置参数出错, 请参考配置说明")
 
     def generate_filename(self):
-        return os.path.join(self.output_file, f"tts-{datetime.now().date()}@{uuid.uuid4().hex}.{self.format}")
+        return os.path.join(self.output_file, f"tts-{datetime.now().date()}@{uuid.uuid4().hex}.{self.audio_file_type}")
 
     async def text_to_speak(self, text, output_file):
         request_params = {}

main/xiaozhi-server/core/providers/tts/doubao.py

@@ -2,15 +2,24 @@ import uuid
 import json
 import base64
 import requests
+
+from config.logger import setup_logging
 from core.utils.util import check_model_key
 from core.providers.tts.base import TTSProviderBase
-from config.logger import setup_logging
+from core.utils.tts import convert_percentage_to_range
+
 
 TAG = __name__
 logger = setup_logging()
 
 
 class TTSProvider(TTSProviderBase):
+    TTS_PARAM_CONFIG = [
+        ("ttsVolume", "volume_ratio", 0.1, 3, 1.0, lambda v: round(float(v), 1)),
+        ("ttsRate", "speed_ratio", 0.2, 3, 1.0, lambda v: round(float(v), 1)),
+        ("ttsPitch", "pitch_ratio", 0.1, 3, 1.0, lambda v: round(float(v), 1)),
+    ]
+
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
         if config.get("appid"):
@@ -34,6 +43,9 @@ class TTSProvider(TTSProviderBase):
         self.volume_ratio = float(volume_ratio) if volume_ratio else 1.0
         self.pitch_ratio = float(pitch_ratio) if pitch_ratio else 1.0
 
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        self._apply_percentage_params(config)
+
         self.api_url = config.get("api_url")
         self.authorization = config.get("authorization")
         self.header = {"Authorization": f"{self.authorization}{self.access_token}"}

main/xiaozhi-server/core/providers/tts/fishspeech.py

@@ -91,7 +91,6 @@ class TTSProvider(TTSProviderBase):
         self.reference_text = parse_string_to_list(
              config.get('ref_text')if config.get('ref_text') else config.get("reference_text")
         )
-        self.format = config.get("response_format", "wav")
         self.audio_file_type = config.get("response_format", "wav")
         self.api_key = config.get("api_key", "YOUR_API_KEY")
         model_key_msg = check_model_key("FishSpeech TTS", self.api_key)
@@ -148,7 +147,7 @@ class TTSProvider(TTSProviderBase):
             ],
             "reference_id": self.reference_id,
             "normalize": self.normalize,
-            "format": self.format,
+            "format": self.audio_file_type,
             "max_new_tokens": self.max_new_tokens,
             "chunk_length": self.chunk_length,
             "top_p": self.top_p,

main/xiaozhi-server/core/providers/tts/huoshan_double_stream.py

@@ -7,11 +7,10 @@ import traceback
 import websockets
 
 from typing import Callable, Any
-from core.utils.tts import MarkdownCleaner
 from config.logger import setup_logging
-from core.utils import opus_encoder_utils
 from core.utils.util import check_model_key
 from core.providers.tts.base import TTSProviderBase
+from core.utils.tts import MarkdownCleaner, convert_percentage_to_range
 from core.providers.tts.dto.dto import SentenceType, ContentType, InterfaceType
 
 
@@ -178,6 +177,22 @@ class TTSProvider(TTSProviderBase):
         self.additions = {**default_additions, **config.get("additions", {})}
         self.mix_speaker = {**default_mix_speaker, **config.get("mix_speaker", {})}
 
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        if "ttsVolume" in config:
+            self.audio_params["loudness_rate"] = int(convert_percentage_to_range(
+                config["ttsVolume"], min_val=-50, max_val=100, base_val=0
+            ))
+
+        if "ttsRate" in config:
+            self.audio_params["speech_rate"] = int(convert_percentage_to_range(
+                config["ttsRate"], min_val=-50, max_val=100, base_val=0
+            ))
+
+        if "ttsPitch" in config:
+            self.additions["post_process"]["pitch"] = int(convert_percentage_to_range(
+                config["ttsPitch"], min_val=-12, max_val=12, base_val=0
+            ))
+
         self.ws_url = config.get("ws_url")
         self.authorization = config.get("authorization")
         self.header = {"Authorization": f"{self.authorization}{self.access_token}"}

main/xiaozhi-server/core/providers/tts/minimax_httpstream.py

@@ -6,12 +6,14 @@ import asyncio
 import aiohttp
 import requests
 import traceback
+
+from core.utils import textUtils
 from config.logger import setup_logging
-from core.utils.tts import MarkdownCleaner
 from core.utils.util import parse_string_to_list
 from core.providers.tts.base import TTSProviderBase
-from core.utils import opus_encoder_utils, textUtils
 from core.providers.tts.dto.dto import SentenceType, ContentType
+from core.utils.tts import MarkdownCleaner, convert_percentage_to_range
+
 
 TAG = __name__
 logger = setup_logging()
@@ -56,6 +58,22 @@ class TTSProvider(TTSProviderBase):
         if self.voice:
             self.voice_setting["voice_id"] = self.voice
 
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        if "ttsVolume" in config:
+            self.voice_setting["vol"] = round(convert_percentage_to_range(
+                config["ttsVolume"], min_val=0.1, max_val=10, base_val=1.0
+            ), 1)
+
+        if "ttsRate" in config:
+            self.voice_setting["speed"] = round(convert_percentage_to_range(
+                config["ttsRate"], min_val=0.5, max_val=2, base_val=1.0
+            ), 1)
+
+        if "ttsPitch" in config:
+            self.voice_setting["pitch"] = int(convert_percentage_to_range(
+                config["ttsPitch"], min_val=-12, max_val=12, base_val=0
+            ))
+
         self.host = "api.minimaxi.com"  # 备用地址：api-bj.minimaxi.com
         self.api_url = f"https://{self.host}/v1/t2a_v2?GroupId={self.group_id}"
         self.header = {

main/xiaozhi-server/core/providers/tts/openai.py

@@ -8,6 +8,10 @@ logger = setup_logging()
 
 
 class TTSProvider(TTSProviderBase):
+    TTS_PARAM_CONFIG = [
+        ("ttsRate", "speed", 0.25, 4, 1, lambda v: round(float(v), 2)),
+    ]
+
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
         self.api_key = config.get("api_key")
@@ -17,13 +21,15 @@ class TTSProvider(TTSProviderBase):
             self.voice = config.get("private_voice")
         else:
             self.voice = config.get("voice", "alloy")
-        self.response_format = config.get("format", "wav")
         self.audio_file_type = config.get("format", "wav")
 
         # 处理空字符串的情况
         speed = config.get("speed", "1.0")
         self.speed = float(speed) if speed else 1.0
 
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        self._apply_percentage_params(config)
+
         self.output_file = config.get("output_dir", "tmp/")
         model_key_msg = check_model_key("TTS", self.api_key)
         if model_key_msg:
@@ -38,7 +44,7 @@ class TTSProvider(TTSProviderBase):
             "model": self.model,
             "input": text,
             "voice": self.voice,
-            "response_format": "wav",
+            "response_format": self.audio_file_type,
             "speed": self.speed,
         }
         response = requests.post(self.api_url, json=data, headers=headers)

main/xiaozhi-server/core/providers/tts/paddle_speech.py

@@ -16,6 +16,11 @@ logger = setup_logging()
 
 
 class TTSProvider(TTSProviderBase):
+    TTS_PARAM_CONFIG = [
+        ("ttsVolume", "volume", 0, 3, 1.0, lambda v: round(float(v), 1)),
+        ("ttsRate", "speed", 0, 3, 1.0, lambda v: round(float(v), 1)),
+    ]
+
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
         self.url = config.get("url", "ws://192.168.1.10:8092/paddlespeech/tts/streaming")
@@ -33,6 +38,10 @@ class TTSProvider(TTSProviderBase):
         self.volume = float(volume) if volume else 1.0
         
         self.delete_audio_file = config.get("delete_audio", True)
+
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        self._apply_percentage_params(config)
+
         if not self.delete_audio_file:
             timestamp = datetime.now().strftime("%Y%m%d_%H%M%S")
             save_path = config.get("save_path")

main/xiaozhi-server/core/providers/tts/siliconflow.py

@@ -3,6 +3,11 @@ from core.providers.tts.base import TTSProviderBase
 
 
 class TTSProvider(TTSProviderBase):
+    TTS_PARAM_CONFIG = [
+        ("ttsVolume", "gain", -10, 10, 0, int),
+        ("ttsRate", "speed", 0.25, 4, 1, lambda v: round(float(v), 2)),
+    ]
+
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
         self.model = config.get("model")
@@ -11,11 +16,13 @@ class TTSProvider(TTSProviderBase):
             self.voice = config.get("private_voice")
         else:
             self.voice = config.get("voice")
-        self.response_format = config.get("response_format", "mp3")
         self.audio_file_type = config.get("response_format", "mp3")
         self.speed = float(config.get("speed", 1.0))
         self.gain = config.get("gain")
 
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        self._apply_percentage_params(config)
+
         self.host = "api.siliconflow.cn"
         self.api_url = f"https://{self.host}/v1/audio/speech"
 
@@ -24,7 +31,7 @@ class TTSProvider(TTSProviderBase):
             "model": self.model,
             "input": text,
             "voice": self.voice,
-            "response_format": self.response_format,
+            "response_format": self.audio_file_type,
         }
         headers = {
             "Authorization": f"Bearer {self.access_token}",

main/xiaozhi-server/core/providers/tts/tencent.py

@@ -10,6 +10,11 @@ from core.providers.tts.base import TTSProviderBase
 
 
 class TTSProvider(TTSProviderBase):
+    TTS_PARAM_CONFIG = [
+        ("ttsVolume", "volume", -10, 10, 0, lambda v: round(float(v), 1)),
+        ("ttsRate", "speed", -2, 6, 0, lambda v: round(float(v), 2)),
+    ]
+
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
         self.appid = config.get("appid")
@@ -24,6 +29,16 @@ class TTSProvider(TTSProviderBase):
         self.output_file = config.get("output_dir")
         self.audio_file_type = config.get("format", "wav")
 
+        # 音频参数配置
+        speed = config.get("speed", "0")
+        self.speed = int(speed) if speed else 0
+
+        volume = config.get("volume", "0")
+        self.volume = int(volume) if volume else 0
+
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        self._apply_percentage_params(config)
+
     def _get_auth_headers(self, request_body):
         """生成鉴权请求头"""
         # 获取当前UTC时间戳
@@ -127,6 +142,10 @@ class TTSProvider(TTSProviderBase):
             "Text": text,  # 合成语音的源文本
             "SessionId": str(uuid.uuid4()),  # 会话ID，随机生成
             "VoiceType": int(self.voice),  # 音色
+            "Codec": self.audio_file_type,  # 音频编码格式
+            "Volume": self.volume,  # 音量
+            "Speed": self.speed,  # 语速
+            "SampleRate": self.conn.sample_rate,  # 采样率部分支持24000
         }
 
         try:

main/xiaozhi-server/core/providers/tts/ttson.py

@@ -12,6 +12,12 @@ logger = setup_logging()
 
 
 class TTSProvider(TTSProviderBase):
+    TTS_PARAM_CONFIG = [
+        ("ttsVolume", "volume_change_dB", -10, 10, 0, int),
+        ("ttsRate", "speed_factor", 0.5, 2, 0, lambda v: round(float(v), 1)),
+        ("ttsPitch", "pitch_factor", -8, 8, 0, lambda v: round(float(v), 1)),
+    ]
+
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
         self.url = config.get(
@@ -29,11 +35,13 @@ class TTSProvider(TTSProviderBase):
         self.stream = str(config.get("stream", False)).lower() in ("true", "1", "yes")
         self.output_file = config.get("output_dir")
         self.pitch_factor = int(config.get("pitch_factor", 0))
-        self.format = config.get("format", "mp3")
         self.audio_file_type = config.get("format", "mp3")
         self.emotion = int(config.get("emotion", 1))
         self.header = {"Content-Type": "application/json"}
 
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        self._apply_percentage_params(config)
+
     def generate_filename(self, extension=".mp3"):
         return os.path.join(
             self.output_file,
@@ -48,7 +56,7 @@ class TTSProvider(TTSProviderBase):
                 "to_lang": self.to_lang,
                 "text": text,
                 "emotion": self.emotion,
-                "format": self.format,
+                "format": self.audio_file_type,
                 "volume_change_dB": self.volume_change_dB,
                 "voice_id": self.voice,
                 "pitch_factor": self.pitch_factor,

main/xiaozhi-server/core/providers/tts/xunfei_stream.py

@@ -9,10 +9,10 @@ import hashlib
 import asyncio
 import traceback
 import websockets
-from typing import Callable, Any
+
 from asyncio import Task
+from typing import Callable, Any
 from config.logger import setup_logging
-from core.utils import opus_encoder_utils
 from core.utils.tts import MarkdownCleaner
 from urllib.parse import urlencode, urlparse
 from core.providers.tts.base import TTSProviderBase
@@ -60,6 +60,12 @@ class XunfeiWSAuth:
 
 
 class TTSProvider(TTSProviderBase):
+    TTS_PARAM_CONFIG = [
+        ("ttsVolume", "volume", 0, 100, 50, int),
+        ("ttsRate", "speed", 0, 100, 50, int),
+        ("ttsPitch", "pitch", 0, 100, 50, int),
+    ]
+
     def __init__(self, config, delete_audio_file):
         super().__init__(config, delete_audio_file)
 
@@ -89,6 +95,9 @@ class TTSProvider(TTSProviderBase):
         pitch = config.get("pitch", "50")
         self.pitch = int(pitch) if pitch else 50
 
+        # 应用百分比调整（如果存在），否则使用公有化配置
+        self._apply_percentage_params(config)
+
         # 音频编码配置
         self.format = config.get("format", "raw")
 

main/xiaozhi-server/core/utils/prompt_manager.py

@@ -251,6 +251,15 @@ class PromptManager:
                         or ""
                     )
 
+            # 获取TTS选择的语言，默认值为中文
+            language = (
+                self.config.get("TTS", {})
+                .get(self.config.get("selected_module", {}).get("TTS", ""), {})
+                .get("language")
+                or "中文"
+            )
+            self.logger.bind(tag=TAG).debug(f"获取到选择的语言: {language}")
+
             # 替换模板变量
             template = Template(self.base_prompt_template)
             enhanced_prompt = template.render(
@@ -265,6 +274,7 @@ class PromptManager:
                 device_id=device_id,
                 client_ip=client_ip,
                 dynamic_context=self.context_data,
+                language=language,
                 *args,
                 **kwargs,
             )

main/xiaozhi-server/core/utils/tts.py

@@ -142,3 +142,29 @@ class MarkdownCleaner:
         text = check_emoji(text)
 
         return text.strip()
+
+def convert_percentage_to_range(percentage, min_val, max_val, base_val=None):
+    """
+    将百分比(-100~100)转换为指定范围的值
+
+    Args:
+        percentage: 百分比值 (-100 到 100)
+        min_val: 目标范围最小值
+        max_val: 目标范围最大值
+        base_val: 基准值（可选，默认为范围中点）
+
+    Returns:
+        转换后的值
+    """
+    percentage, min_val, max_val = float(percentage), float(min_val), float(max_val)
+    base_val = float(base_val) if base_val is not None else (min_val + max_val) / 2
+
+    if percentage < 0:
+        # 负百分比：从 base_val 向 min_val 线性插值
+        result = base_val + (base_val - min_val) * (percentage / 100)
+    else:
+        # 正百分比：从 base_val 向 max_val 线性插值
+        result = base_val + (max_val - base_val) * (percentage / 100)
+
+    # 确保结果在有效范围内
+    return max(min_val, min(max_val, result))