ChatTTS

GitHub：2noise/ChatTTS: A generative speech model for daily dialogue.。他是这样介绍的：一款适用于日常对话的生成式语音模型。

Windows

下载项目

2024-12-26这里下载的是v0.2.1，如下图：

ChatTTS 项目，下载版本是 v0.2.1，解压位置

虚拟环境/安装依赖

conda create -n chattts python=3.11
conda activate chattts
pip install -r requirements.txt

点击查看详情

ChatTTS 项目，conda 创建虚拟环境

ChatTTS 项目，激活虚拟环境、安装依赖

ChatTTS 项目，安装依赖成功

启动项目 - WebUI

我去，好顺利啊！这里直接启动带 Web 页面，对 Web 有天生的好感……

python examples/web/webui.py

这一启动可不得了啊，这一拉流日志都看不过来啊！自己偷偷下载了不少东西，请看：

ChatTTS 项目，启动项目：python examples/web/webui.py 。第一次启动，下载依赖模型

ChatTTS 项目，第一次启动 WebUI 项目时，下载的依赖模型

人家下载就下载呗，人家最后还不是跑起来了，你看：

ChatTTS 项目，启动 WebUI 项目成功，浏览器页面截图

随后测试了一下，很顺利，看一下效果截图吧：

ChatTTS 项目，启动 WebUI 项目之后，生成语音成功

启动项目 - API

这才是咱最想要的，有了她，才能和其他的应用结合在一起。当然，项目还是很贴心的，提供了 API 启动方式：examples/api。该目录下有一个requirements.txt文件，仅包含了两个依赖：fastapi、requests，咱又给他添加了一个 uvicorn 。之后在安装一下pip install -r examples/api/requirements.txt，如下图：

ChatTTS 项目，启动 API 项目，安装依赖

安装成功之后就可以启动了python examples/api/main.py，来了，来了，怎么就不能了让人省心呢？

ChatTTS 项目，启动 API 项目，首次启动报错

改动了好几处地方，这里就直接全部弄过来了，如下：

点击查看详情

import os
import sys
now_dir = os.getcwd()
sys.path.append(now_dir)

from pydantic import BaseModel
import torch
from tools.logger import get_logger
from tools.audio import pcm_arr_to_mp3_view
import ChatTTS
from typing import Optional
import io
import zipfile

import uvicorn
from fastapi import FastAPI
from fastapi.responses import StreamingResponse
from contextlib import asynccontextmanager


if sys.platform == "darwin":
    os.environ["PYTORCH_ENABLE_MPS_FALLBACK"] = "1"

logger = get_logger("Command")


@asynccontextmanager
async def lifespan(app: FastAPI):
    global chat

    chat = ChatTTS.Chat(get_logger("ChatTTS"))
    logger.info("Initializing ChatTTS...")
    if chat.load():
        logger.info("Models loaded successfully.")
    else:
        logger.error("Models load failed.")
        sys.exit(1)
    yield

app = FastAPI(lifespan=lifespan)

class ChatTTSParams(BaseModel):
    text: list[str]
    stream: bool = False
    lang: Optional[str] = None
    skip_refine_text: bool = False
    refine_text_only: bool = False
    use_decoder: bool = True
    do_text_normalization: bool = True
    do_homophone_replacement: bool = False
    params_refine_text: ChatTTS.Chat.RefineTextParams
    params_infer_code: ChatTTS.Chat.InferCodeParams

@app.post("/generate_voice")
async def generate_voice(params: ChatTTSParams):
    logger.info("Text input: %s", str(params.text))

    # audio seed
    if params.params_infer_code.manual_seed is not None:
        torch.manual_seed(params.params_infer_code.manual_seed)
        params.params_infer_code.spk_emb = chat.sample_random_speaker()

    # text seed for text refining
    if params.params_refine_text:
        text = chat.infer(
            text=params.text, skip_refine_text=False, refine_text_only=True
        )
        logger.info(f"Refined text: {text}")
    else:
        # no text refining
        text = params.text

    logger.info("Use speaker:")
    logger.info(params.params_infer_code.spk_emb)

    logger.info("Start voice inference.")
    wavs = chat.infer(
        text=text,
        stream=params.stream,
        lang=params.lang,
        skip_refine_text=params.skip_refine_text,
        use_decoder=params.use_decoder,
        do_text_normalization=params.do_text_normalization,
        do_homophone_replacement=params.do_homophone_replacement,
        params_infer_code=params.params_infer_code,
        params_refine_text=params.params_refine_text,
    )
    logger.info("Inference completed.")

    # zip all of the audio files together
    buf = io.BytesIO()
    with zipfile.ZipFile(
        buf, "a", compression=zipfile.ZIP_DEFLATED, allowZip64=False
    ) as f:
        for idx, wav in enumerate(wavs):
            f.writestr(f"{idx}.mp3", pcm_arr_to_mp3_view(wav))
    logger.info("Audio generation successful.")
    buf.seek(0)

    response = StreamingResponse(buf, media_type="application/zip")
    response.headers["Content-Disposition"] = "attachment; filename=audio_files.zip"
    return response

if __name__ == "__main__":
    uvicorn.run(app, host="0.0.0.0", port=13140)

之后就可以成功启动了，如下图：

ChatTTS 项目，启动 API 项目成功，浏览器 API 接口文档

点击查看具体的依赖版本

#	requirements.txt
numpy==1.26.4
numba==0.60.0
torch==2.5.1
torchaudio==2.5.1
tqdm==4.67.1
vector_quantize_pytorch
transformers==4.47.1
vocos==0.1.0
IPython==8.31.0
gradio==5.9.1
pybase16384==0.3.7
pynini==2.1.5; sys_platform == 'linux'
WeTextProcessing; sys_platform == 'linux'
nemo_text_processing; sys_platform == 'linux'
av==14.0.1
pydub==0.25.1

#	examples/api/requirements.txt
fastapi==0.115.6
requests==2.32.3
uvicorn==0.34.0

心想：现在总行了吧！你想多了，还是有问题啊，如下：

点击查看错误堆栈

INFO:     127.0.0.1:55969 - "POST /generate_voice HTTP/1.1" 500 Internal Server Error
ERROR:    Exception in ASGI application
Traceback (most recent call last):
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\uvicorn\protocols\http\h11_impl.py", line 403, in run_asgi
    result = await app(  # type: ignore[func-returns-value]
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\uvicorn\middleware\proxy_headers.py", line 60, in __call__
    return await self.app(scope, receive, send)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\fastapi\applications.py", line 1054, in __call__
    await super().__call__(scope, receive, send)
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\applications.py", line 113, in __call__
    await self.middleware_stack(scope, receive, send)
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\middleware\errors.py", line 187, in __call__
    raise exc
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\middleware\errors.py", line 165, in __call__
    await self.app(scope, receive, _send)
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\middleware\exceptions.py", line 62, in __call__
    await wrap_app_handling_exceptions(self.app, conn)(scope, receive, send)
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\_exception_handler.py", line 53, in wrapped_app
    raise exc
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\_exception_handler.py", line 42, in wrapped_app
    await app(scope, receive, sender)
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\routing.py", line 715, in __call__
    await self.middleware_stack(scope, receive, send)
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\routing.py", line 735, in app
    await route.handle(scope, receive, send)
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\routing.py", line 288, in handle
    await self.app(scope, receive, send)
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\routing.py", line 76, in app
    await wrap_app_handling_exceptions(app, request)(scope, receive, send)
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\_exception_handler.py", line 53, in wrapped_app
    raise exc
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\_exception_handler.py", line 42, in wrapped_app
    await app(scope, receive, sender)
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\starlette\routing.py", line 73, in app
    response = await f(request)
               ^^^^^^^^^^^^^^^^
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\fastapi\routing.py", line 301, in app
    raw_response = await run_endpoint_function(
                   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\fastapi\routing.py", line 212, in run_endpoint_function
    return await dependant.call(**values)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "F:\mllm\ChatTTS-0.2.1\examples\api\main.py", line 92, in generate_voice
    wavs = chat.infer(
           ^^^^^^^^^^^
  File "F:\mllm\ChatTTS-0.2.1\ChatTTS\core.py", line 221, in infer
    return next(res_gen)
           ^^^^^^^^^^^^^
  File "F:\mllm\ChatTTS-0.2.1\ChatTTS\core.py", line 385, in _infer
    for result in self._infer_code(
                  ^^^^^^^^^^^^^^^^^
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\torch\utils\_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "F:\mllm\ChatTTS-0.2.1\ChatTTS\core.py", line 485, in _infer_code
    self.speaker.decode_prompt(params.spk_smp)
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\site-packages\torch\utils\_contextlib.py", line 116, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "F:\mllm\ChatTTS-0.2.1\ChatTTS\model\speaker.py", line 112, in decode_prompt
    lzma.decompress(
  File "C:\Users\xiaodu\.conda\envs\chattts\Lib\lzma.py", line 343, in decompress
    res = decomp.decompress(data)
          ^^^^^^^^^^^^^^^^^^^^^^^
_lzma.LZMAError: Corrupt input data

这个错误发现是chat.infer(...)方法内部出了问题，感觉应该是参数的问题，因为 WebUI 项目也会调用这个方法，有时间对比一下这两处调用传入的参数有何不同

Python 包

我去，其他的都是浮云，这里才是重点，都有 Python 包了，还要啥 API 啊，集成更方便了👍👍👍。官网的“开发教程”章节给出了使用方法：

ChatTTS 项目，开发教程，Python 包使用方法