部分TTS参数调整

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

@@ -536,3 +536,10 @@ databaseChangeLog:
         - sqlFile:
             encoding: utf8
             path: classpath:db/changelog/202602061650.sql
+  - changeSet:
+      id: 202602271137
+      author: RanChen
+      changes:
+        - sqlFile:
+            encoding: utf8
+            path: classpath:db/changelog/202602271137.sql

main/xiaozhi-server/config.yaml

@@ -813,6 +813,9 @@ TTS:
     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
@@ -984,7 +987,10 @@ TTS:
     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

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

@@ -100,7 +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"):
@@ -179,7 +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

@@ -103,7 +103,6 @@ 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"):

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/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/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/siliconflow.py

@@ -5,7 +5,7 @@ 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), 1)),
+        ("ttsRate", "speed", 0.25, 4, 1, lambda v: round(float(v), 2)),
     ]
 
     def __init__(self, config, delete_audio_file):
@@ -16,7 +16,6 @@ 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")
@@ -32,7 +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

@@ -35,7 +35,6 @@ 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"}
@@ -57,7 +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/utils/tts.py

@@ -156,10 +156,9 @@ def convert_percentage_to_range(percentage, min_val, max_val, base_val=None):
     Returns:
         转换后的值
     """
-    if base_val is None:
-        base_val = (min_val + max_val) / 2
+    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
 
-    # 百分比 -100 对应 min_val, 0 对应 base_val, 100 对应 max_val
     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)