test1

Dockerfile

@@ -0,0 +1,44 @@
+# 构建阶段
+FROM python:3.12-slim as builder
+
+# 设置工作目录
+WORKDIR /app
+
+# 安装系统依赖
+RUN apt-get update && apt-get install -y --no-install-recommends \
+    build-essential \
+    && rm -rf /var/lib/apt/lists/*
+
+# 复制依赖文件
+COPY requirements.txt .
+
+# 安装Python依赖
+RUN pip install --no-cache-dir -r requirements.txt
+
+# 运行阶段
+FROM python:3.12-slim
+
+# 创建非特权用户
+RUN useradd -m -s /bin/bash app
+
+# 设置工作目录
+WORKDIR /app
+
+# 复制应用代码和依赖
+COPY --from=builder /usr/local/lib/python3.12/site-packages /usr/local/lib/python3.12/site-packages
+COPY . .
+
+# 设置权限
+RUN chown -R app:app /app
+
+# 切换到非特权用户
+USER app
+
+# 设置环境变量
+ENV PYTHONUNBUFFERED=1
+
+# 暴露端口
+EXPOSE 8000
+
+# 启动命令
+CMD ["python", "-m", "uvicorn", "api:app", "--host", "0.0.0.0", "--port", "8000"]

api.py

@@ -0,0 +1,256 @@
+# coding=utf-8
+
+from io import BytesIO
+from typing import Optional
+
+from fastapi import FastAPI, File, UploadFile, HTTPException, Depends
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.security import HTTPBearer, HTTPAuthorizationCredentials
+import numpy as np
+import torch
+import torchaudio
+from funasr import AutoModel
+from dotenv import load_dotenv
+import os
+
+# 加载环境变量
+load_dotenv()
+
+# 获取API Token
+API_TOKEN = os.getenv("API_TOKEN")
+if not API_TOKEN:
+    raise RuntimeError("API_TOKEN environment variable is not set")
+
+# 设置认证
+security = HTTPBearer()
+
+app = FastAPI(
+    title="SenseVoice API",
+    description="语音识别 API 服务",
+    version="1.0.0"
+)
+
+# 允许跨域请求
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+# 初始化模型
+model = AutoModel(
+    model="FunAudioLLM/SenseVoiceSmall",
+    vad_model="iic/speech_fsmn_vad_zh-cn-16k-common-pytorch",
+    vad_kwargs={"max_single_segment_time": 30000},
+    hub="hf",
+    device="cuda"
+)
+
+# 复用原有的格式化函数
+emo_dict = {
+    "<|HAPPY|>": "😊",
+    "<|SAD|>": "😔",
+    "<|ANGRY|>": "😡",
+    "<|NEUTRAL|>": "",
+    "<|FEARFUL|>": "😰",
+    "<|DISGUSTED|>": "🤢",
+    "<|SURPRISED|>": "😮",
+}
+
+event_dict = {
+    "<|BGM|>": "🎼",
+    "<|Speech|>": "",
+    "<|Applause|>": "👏",
+    "<|Laughter|>": "😀",
+    "<|Cry|>": "😭",
+    "<|Sneeze|>": "🤧",
+    "<|Breath|>": "",
+    "<|Cough|>": "🤧",
+}
+
+emoji_dict = {
+    "<|nospeech|><|Event_UNK|>": "❓",
+    "<|zh|>": "",
+    "<|en|>": "",
+    "<|yue|>": "",
+    "<|ja|>": "",
+    "<|ko|>": "",
+    "<|nospeech|>": "",
+    "<|HAPPY|>": "😊",
+    "<|SAD|>": "😔",
+    "<|ANGRY|>": "😡",
+    "<|NEUTRAL|>": "",
+    "<|BGM|>": "🎼",
+    "<|Speech|>": "",
+    "<|Applause|>": "👏",
+    "<|Laughter|>": "😀",
+    "<|FEARFUL|>": "😰",
+    "<|DISGUSTED|>": "🤢",
+    "<|SURPRISED|>": "😮",
+    "<|Cry|>": "😭",
+    "<|EMO_UNKNOWN|>": "",
+    "<|Sneeze|>": "🤧",
+    "<|Breath|>": "",
+    "<|Cough|>": "😷",
+    "<|Sing|>": "",
+    "<|Speech_Noise|>": "",
+    "<|withitn|>": "",
+    "<|woitn|>": "",
+    "<|GBG|>": "",
+    "<|Event_UNK|>": "",
+}
+
+lang_dict = {
+    "<|zh|>": "<|lang|>",
+    "<|en|>": "<|lang|>",
+    "<|yue|>": "<|lang|>",
+    "<|ja|>": "<|lang|>",
+    "<|ko|>": "<|lang|>",
+    "<|nospeech|>": "<|lang|>",
+}
+
+emo_set = {"😊", "😔", "😡", "😰", "🤢", "😮"}
+event_set = {"🎼", "👏", "😀", "😭", "🤧", "😷"}
+
+
+def format_str(s):
+    for sptk in emoji_dict:
+        s = s.replace(sptk, emoji_dict[sptk])
+    return s
+
+
+def format_str_v2(s):
+    sptk_dict = {}
+    for sptk in emoji_dict:
+        sptk_dict[sptk] = s.count(sptk)
+        s = s.replace(sptk, "")
+    emo = "<|NEUTRAL|>"
+    for e in emo_dict:
+        if sptk_dict[e] > sptk_dict[emo]:
+            emo = e
+    for e in event_dict:
+        if sptk_dict[e] > 0:
+            s = event_dict[e] + s
+    s = s + emo_dict[emo]
+
+    for emoji in emo_set.union(event_set):
+        s = s.replace(" " + emoji, emoji)
+        s = s.replace(emoji + " ", emoji)
+    return s.strip()
+
+
+def format_str_v3(s):
+    def get_emo(s):
+        return s[-1] if s[-1] in emo_set else None
+
+    def get_event(s):
+        return s[0] if s[0] in event_set else None
+
+    s = s.replace("<|nospeech|><|Event_UNK|>", "❓")
+    for lang in lang_dict:
+        s = s.replace(lang, "<|lang|>")
+    s_list = [format_str_v2(s_i).strip(" ") for s_i in s.split("<|lang|>")]
+    new_s = " " + s_list[0]
+    cur_ent_event = get_event(new_s)
+    for i in range(1, len(s_list)):
+        if len(s_list[i]) == 0:
+            continue
+        if get_event(s_list[i]) == cur_ent_event and get_event(s_list[i]) != None:
+            s_list[i] = s_list[i][1:]
+        cur_ent_event = get_event(s_list[i])
+        if get_emo(s_list[i]) != None and get_emo(s_list[i]) == get_emo(new_s):
+            new_s = new_s[:-1]
+        new_s += s_list[i].strip().lstrip()
+    new_s = new_s.replace("The.", " ")
+    return new_s.strip()
+
+
+async def process_audio(audio_data: bytes, language: str = "auto") -> str:
+    """处理音频数据并返回识别结果"""
+    try:
+        # 将字节数据转换为 numpy 数组
+        audio_buffer = BytesIO(audio_data)
+        waveform, sample_rate = torchaudio.load(audio_buffer)
+        
+        # 转换为单声道
+        if waveform.shape[0] > 1:
+            waveform = waveform.mean(dim=0)
+        
+        # 转换为 numpy array 并归一化
+        input_wav = waveform.numpy().astype(np.float32)
+        
+        # 重采样到 16kHz
+        if sample_rate != 16000:
+            resampler = torchaudio.transforms.Resample(sample_rate, 16000)
+            input_wav = resampler(torch.from_numpy(input_wav)[None, :])[0, :].numpy()
+        
+        # 模型推理
+        text = model.generate(
+            input=input_wav,
+            cache={},
+            language=language,
+            use_itn=True,
+            batch_size_s=500,
+            merge_vad=True
+        )
+        
+        # 格式化结果
+        result = text[0]["text"]
+        result = format_str_v3(result)
+        
+        return result
+    
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=f"音频处理失败：{str(e)}")
+
+
+async def verify_token(credentials: HTTPAuthorizationCredentials = Depends(security)):
+    """验证Bearer Token"""
+    if credentials.credentials != API_TOKEN:
+        raise HTTPException(
+            status_code=401,
+            detail="Invalid authentication token",
+            headers={"WWW-Authenticate": "Bearer"}
+        )
+    return credentials
+
+@app.post("/v1/audio/transcriptions")
+async def transcribe_audio(
+    file: UploadFile = File(...),
+    model: Optional[str] = "FunAudioLLM/SenseVoiceSmall",
+    language: Optional[str] = "auto",
+    token: HTTPAuthorizationCredentials = Depends(verify_token)
+):
+    """音频转写接口
+    
+    Args:
+        file: 音频文件（支持常见音频格式）
+        model: 模型名称，目前仅支持 FunAudioLLM/SenseVoiceSmall
+        language: 语言代码，支持 auto/zh/en/yue/ja/ko/nospeech
+    
+    Returns:
+        {"text": "识别结果"}
+    """
+    if not file.filename.lower().endswith((".mp3", ".wav", ".flac", ".ogg", ".m4a")):
+        raise HTTPException(status_code=400, detail="不支持的音频格式")
+    
+    if model != "FunAudioLLM/SenseVoiceSmall":
+        raise HTTPException(status_code=400, detail="不支持的模型")
+    
+    if language not in ["auto", "zh", "en", "yue", "ja", "ko", "nospeech"]:
+        raise HTTPException(status_code=400, detail="不支持的语言")
+    
+    try:
+        content = await file.read()
+        text = await process_audio(content, language)
+        return {"text": text}
+    
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+if __name__ == "__main__":
+    import uvicorn
+    uvicorn.run(app, host="0.0.0.0", port=8000)

build.sh

@@ -0,0 +1,30 @@
+#!/bin/bash
+
+set -ex
+
+# 设置变量
+IMAGE_NAME="sensevoice-api"
+CONTAINER_NAME="sensevoice-api-container"
+PORT=8000
+
+# 停止并删除已存在的同名容器
+if [ "$(docker ps -aq -f name=$CONTAINER_NAME)" ]; then
+    echo "停止并删除已存在的容器..."
+    docker stop $CONTAINER_NAME
+    docker rm $CONTAINER_NAME
+fi
+
+# 构建Docker镜像
+echo "构建Docker镜像..."
+docker build -t $IMAGE_NAME .
+
+# 运行容器
+echo "启动容器..."
+docker run -d \
+    --name $CONTAINER_NAME \
+    -p $PORT:8000 \
+    -e API_TOKEN="your-secret-token-here" \
+    -e PYTHONUNBUFFERED=1 \
+    $IMAGE_NAME
+
+echo "容器启动成功！服务运行在 http://localhost:$PORT"

requirements.txt

@@ -0,0 +1,15 @@
+# Web 框架和服务器
+fastapi==0.104.1  # 现代、快速的 Web API 框架
+uvicorn==0.24.0  # 轻量级 ASGI 服务器
+python-multipart==0.0.6  # 处理文件上传的multipart表单数据
+
+# 数据处理和科学计算
+numpy==1.26.2  # 科学计算基础库，提供多维数组支持
+
+# 深度学习和音频处理
+torch==2.1.1  # PyTorch深度学习框架
+torchaudio==2.1.1  # PyTorch音频处理库
+funasr==0.8.1  # 语音识别模型库
+
+# 工具库
+python-dotenv==1.0.0  # 从.env文件加载环境变量

start.sh

@@ -0,0 +1,7 @@
+#!/bin/bash
+
+# 设置API认证Token
+export API_TOKEN="your-secret-token-here"
+
+# 启动FastAPI服务
+python -m uvicorn api:app --host 0.0.0.0 --port 8000