mirror of
https://github.com/xinnan-tech/xiaozhi-esp32-server.git
synced 2026-07-25 16:43:55 +08:00
add:增加智控台模型配置各个模型的使用说明
This commit is contained in:
hrz
@@ -210,8 +210,14 @@ ASR:
|
||||
model_dir: models/SenseVoiceSmall
|
||||
output_dir: tmp/
|
||||
FunASRServer:
|
||||
# 支持FunASR服务，部署方法：https://github.com/modelscope/FunASR/blob/main/runtime/docs/SDK_advanced_guide_online_zh.md
|
||||
# mode 预置 offline
|
||||
# 独立部署FunASR，使用FunASR的API服务，只需要五句话
|
||||
# 第一句：mkdir -p ./funasr-runtime-resources/models
|
||||
# 第二句：sudo docker run -d -p 10096:10095 --privileged=true -v $PWD/funasr-runtime-resources/models:/workspace/models registry.cn-hangzhou.aliyuncs.com/funasr_repo/funasr:funasr-runtime-sdk-online-cpu-0.1.12
|
||||
# 上一句话执行后会进入到容器，继续第三句：cd FunASR/runtime
|
||||
# 不要退出容器，继续在容器中执行第四句：nohup bash run_server_2pass.sh --download-model-dir /workspace/models --vad-dir damo/speech_fsmn_vad_zh-cn-16k-common-onnx --model-dir damo/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k-common-vocab8404-onnx --online-model-dir damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-online-onnx --punc-dir damo/punc_ct-transformer_zh-cn-common-vad_realtime-vocab272727-onnx --lm-dir damo/speech_ngram_lm_zh-cn-ai-wesp-fst --itn-dir thuduj12/fst_itn_zh --hotword /workspace/models/hotwords.txt > log.txt 2>&1 &
|
||||
# 上一句话执行后会进入到容器，继续第五句：tail -f log.txt
|
||||
# 第五句话执行完后，会看到模型下载日志，下载完后就可以连接使用了
|
||||
# 以上是使用CPU推理，如果有GPU，详细参考：https://github.com/modelscope/FunASR/blob/main/runtime/docs/SDK_advanced_guide_online_zh.md
|
||||
type: fun_server
|
||||
host: 127.0.0.1
|
||||
port: 10096
|
||||
@@ -272,12 +278,12 @@ LLM:
|
||||
DoubaoLLM:
|
||||
# 定义LLM API类型
|
||||
type: openai
|
||||
# 先开通服务，打开以下网址，开通的服务搜索Doubao-pro-32k，开通它
|
||||
# 先开通服务，打开以下网址，开通的服务搜索Doubao-1.5-pro，开通它
|
||||
# 开通改地址：https://console.volcengine.com/ark/region:ark+cn-beijing/openManagement?LLM=%7B%7D&OpenTokenDrawer=false
|
||||
# 免费额度500000token
|
||||
# 开通后，进入这里获取密钥：https://console.volcengine.com/ark/region:ark+cn-beijing/apiKey?apikey=%7B%7D
|
||||
base_url: https://ark.cn-beijing.volces.com/api/v3
|
||||
model_name: doubao-pro-32k-functioncall-241028
|
||||
model_name: doubao-1-5-pro-32k-250115
|
||||
api_key: 你的doubao web key
|
||||
DeepSeekLLM:
|
||||
# 定义LLM API类型
|
||||
@@ -453,12 +459,12 @@ TTS:
|
||||
GPT_SOVITS_V2:
|
||||
# 定义TTS API类型
|
||||
#启动tts方法：
|
||||
#python api_v2.py -a 127.0.0.1 -p 9880 -c GPT_SoVITS/configs/caixukun.yaml
|
||||
#python api_v2.py -a 127.0.0.1 -p 9880 -c GPT_SoVITS/configs/demo.yaml
|
||||
type: gpt_sovits_v2
|
||||
url: "http://127.0.0.1:9880/tts"
|
||||
output_dir: tmp/
|
||||
text_lang: "auto"
|
||||
ref_audio_path: "caixukun.wav"
|
||||
ref_audio_path: "demo.wav"
|
||||
prompt_text: ""
|
||||
prompt_lang: "zh"
|
||||
top_k: 5
|
||||
@@ -591,8 +597,8 @@ TTS:
|
||||
ACGNTTS:
|
||||
#在线网址：https://acgn.ttson.cn/
|
||||
#token购买：www.ttson.cn
|
||||
#开发相关疑问请提交至3497689533@qq.com
|
||||
#角色id获取地址：ctrl+f快速检索角色——网站管理者不允许发布,可询问网站管理者：1069379506
|
||||
#开发相关疑问请提交至网站上的qq
|
||||
#角色id获取地址：ctrl+f快速检索角色——网站管理者不允许发布,可询问网站管理者
|
||||
#各参数意义见开发文档：https://www.yuque.com/alexuh/skmti9/wm6taqislegb02gd?singleDoc#
|
||||
type: ttson
|
||||
token: your_token
|
||||
|
||||
@@ -43,11 +43,11 @@ class ASRProvider(ASRProviderBase):
|
||||
|
||||
# 确保输出目录存在
|
||||
os.makedirs(self.output_dir, exist_ok=True)
|
||||
|
||||
|
||||
# 初始化模型文件路径
|
||||
model_files = {
|
||||
"model.int8.onnx": os.path.join(self.model_dir, "model.int8.onnx"),
|
||||
"tokens.txt": os.path.join(self.model_dir, "tokens.txt")
|
||||
"tokens.txt": os.path.join(self.model_dir, "tokens.txt"),
|
||||
}
|
||||
|
||||
# 下载并检查模型文件
|
||||
@@ -58,15 +58,15 @@ class ASRProvider(ASRProviderBase):
|
||||
model_file_download(
|
||||
model_id="pengzhendong/sherpa-onnx-sense-voice-zh-en-ja-ko-yue",
|
||||
file_path=file_name,
|
||||
local_dir=self.model_dir
|
||||
local_dir=self.model_dir,
|
||||
)
|
||||
|
||||
|
||||
if not os.path.isfile(file_path):
|
||||
raise FileNotFoundError(f"模型文件下载失败: {file_path}")
|
||||
|
||||
|
||||
self.model_path = model_files["model.int8.onnx"]
|
||||
self.tokens_path = model_files["tokens.txt"]
|
||||
|
||||
|
||||
except Exception as e:
|
||||
logger.bind(tag=TAG).error(f"模型文件处理失败: {str(e)}")
|
||||
raise
|
||||
@@ -83,7 +83,7 @@ class ASRProvider(ASRProviderBase):
|
||||
use_itn=True,
|
||||
)
|
||||
|
||||
def save_audio_to_file(self, opus_data: List[bytes], session_id: str) -> str:
|
||||
def save_audio_to_file(self, pcm_data: List[bytes], session_id: str) -> str:
|
||||
"""PCM数据保存为WAV文件"""
|
||||
file_name = f"asr_{session_id}_{uuid.uuid4()}.wav"
|
||||
file_path = os.path.join(self.output_dir, file_name)
|
||||
@@ -135,7 +135,9 @@ class ASRProvider(ASRProviderBase):
|
||||
samples_float32 = samples_float32 / 32768
|
||||
return samples_float32, f.getframerate()
|
||||
|
||||
async def speech_to_text(self, opus_data: List[bytes], session_id: str) -> Tuple[Optional[str], Optional[str]]:
|
||||
async def speech_to_text(
|
||||
self, opus_data: List[bytes], session_id: str
|
||||
) -> Tuple[Optional[str], Optional[str]]:
|
||||
"""语音转文本主处理逻辑"""
|
||||
file_path = None
|
||||
try:
|
||||
@@ -143,7 +145,9 @@ class ASRProvider(ASRProviderBase):
|
||||
start_time = time.time()
|
||||
pcm_data = self.decode_opus(opus_data, session_id)
|
||||
file_path = self.save_audio_to_file(pcm_data, session_id)
|
||||
logger.bind(tag=TAG).debug(f"音频文件保存耗时: {time.time() - start_time:.3f}s | 路径: {file_path}")
|
||||
logger.bind(tag=TAG).debug(
|
||||
f"音频文件保存耗时: {time.time() - start_time:.3f}s | 路径: {file_path}"
|
||||
)
|
||||
|
||||
# 语音识别
|
||||
start_time = time.time()
|
||||
@@ -152,14 +156,15 @@ class ASRProvider(ASRProviderBase):
|
||||
s.accept_waveform(sample_rate, samples)
|
||||
self.model.decode_stream(s)
|
||||
text = s.result.text
|
||||
logger.bind(tag=TAG).debug(f"语音识别耗时: {time.time() - start_time:.3f}s | 结果: {text}")
|
||||
logger.bind(tag=TAG).debug(
|
||||
f"语音识别耗时: {time.time() - start_time:.3f}s | 结果: {text}"
|
||||
)
|
||||
|
||||
return text, file_path
|
||||
|
||||
except Exception as e:
|
||||
logger.bind(tag=TAG).error(f"语音识别失败: {e}", exc_info=True)
|
||||
return "", None
|
||||
|
||||
finally:
|
||||
# 文件清理逻辑
|
||||
if self.delete_audio_file and file_path and os.path.exists(file_path):
|
||||
|
||||
Reference in New Issue
Block a user