本地进行语音文字互转

文字转语音 ChatTTS

https://github.com/2noise/ChatTTS

• 用下面的代码即可实现，输出一个wav音频文件，即为转化的结果

from torch import manual_seed

import ChatTTS
import torch
import torchaudio

chat = ChatTTS.Chat()
chat.load(compile=True) # Set to True for better performance

texts = ["今年上半年，各地因地制宜，加快建设各具特色的产业集群。"]

# 设置种子, 保持音色稳定
torch.manual_seed(120)

wavs = chat.infer(texts)

for i in range(len(wavs)):
    """
    In some versions of torchaudio, the first line works but in other versions, so does the second line.
    """
    try:
        torchaudio.save(f"basic_output{i}.wav", torch.from_numpy(wavs[i]).unsqueeze(0), 24000, format="wav")
    except:
        torchaudio.save(f"basic_output{i}.wav", torch.from_numpy(wavs[i]), 24000, format="wav")

• 需要安装soundfile来处理音频格式

pip install soundfile

RuntimeError: narrow(): length must be non-negative.

• transformer版本问题导致的，运行下面命令即可解决

pip install transformers==4.53.2

https://github.com/2noise/ChatTTS/issues/955

语音转文字 whisper

https://github.com/openai/whisper

• 把上面转出来的wav文件，用whisper再转成文字，按照官方代码实例即可