sunx.ai

PyTorch  2.1.0

Python  3.10(ubuntu22.04)

Cuda  12.1

git clone https://github.com/2noise/ChatTTS.git
cd ChatTTS
pip install -r requirements.txt


#如果要啟動webui.py 就多安裝一下gradio
pip install gradio

python webui.py
#執行後他會自己下載模型到huggingface文件夾，需要自備魔法，或者用huggingface代理網站
#指定端口執行
python webui.py --server_port 1234

#簡單版 api.py      執行 python api.py
import ChatTTS
import torch
import torchaudio
from IPython.display import Audio

# 初始化 ChatTTS 模型
chat = ChatTTS.Chat()
chat.load_models(compile=False)  # 如果需要更高性能，可以設置為 True
rand_spk = chat.sample_random_speaker()
# 要轉換的文本
texts = ["你好，我是文字轉語音"]

# 執行推理
wavs = chat.infer(texts)

# 保存音頻文件
torchaudio.save("output1.wav", torch.from_numpy(wavs[0]), 24000)

###################################
# 高級班 v-api.py       執行 python v-api.py
import ChatTTS 
import torch
import torchaudio
from IPython.display import Audio


chat = ChatTTS.Chat()
chat.load_models(compile=False)  # 如果需要更高性能，可以設置為 True

###################################
# Sample a speaker from Gaussian.

rand_spk = chat.sample_random_speaker()

params_infer_code = {
  'spk_emb': rand_spk, # add sampled speaker 
  'temperature': .3, # using custom temperature
  'top_P': 0.7, # top P decode
  'top_K': 20, # top K decode
}

###################################
# For sentence level manual control.

# use oral_(0-9), laugh_(0-2), break_(0-7) 
# to generate special token in text to synthesize.
params_refine_text = {
  'prompt': '[oral_2][laugh_0][break_6]'
} 

###################################
# For word level manual control.
text = 'What is [uv_break]your favorite english food?[laugh][lbreak]'
wav = chat.infer(text, skip_refine_text=True, params_refine_text=params_refine_text,  params_infer_code=params_infer_code)
torchaudio.save("output2.wav", torch.from_numpy(wav[0]), 24000)