使用3s极速复刻时报错

[keyiis]

我是用官方文档的命令启动webui python3 webui.py --port 50000 --model_dir pretrained_models/CosyVoice-300M
启动后界面可以正常打开，如下

上面是我的配置，点击运行后提示错误，终端输出如下：

2024-10-28 11:15:36,546 INFO get zero_shot inference request
  0%|                                                                                                                                                                                          | 0/1 [00:00<?, ?it/s]2024-10-28 11:15:55,785 INFO synthesis text 我是通义实验室语音团队全新推出的生成式语音大模型,提供舒适自然的语音合成能力。
Exception in thread Thread-7:
Traceback (most recent call last):
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\threading.py", line 932, in _bootstrap_inner
    self.run()
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\threading.py", line 870, in run
    self._target(*self._args, **self._kwargs)
  File "D:\projects\ai\CosyVoice\cosyvoice\cli\model.py", line 93, in llm_job
    for i in self.llm.inference(text=text.to(self.device),
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\torch\utils\_contextlib.py", line 35, in generator_context
    response = gen.send(None)
  File "D:\projects\ai\CosyVoice\cosyvoice\llm\llm.py", line 172, in inference
    text, text_len = self.encode(text, text_len)
  File "D:\projects\ai\CosyVoice\cosyvoice\llm\llm.py", line 75, in encode
    encoder_out, encoder_mask = self.text_encoder(text, text_lengths, decoding_chunk_size=1, num_decoding_left_chunks=-1)
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript, serialized code (most recent call last):
  File "code/__torch__/cosyvoice/transformer/encoder/___torch_mangle_5.py", line 22, in forward
    masks = torch.bitwise_not(torch.unsqueeze(mask, 1))
    embed = self.embed
    _0 = torch.add(torch.matmul(xs, CONSTANTS.c0), CONSTANTS.c1)
                   ~~~~~~~~~~~~ <--- HERE
    input = torch.layer_norm(_0, [1024], CONSTANTS.c2, CONSTANTS.c3)
    pos_enc = embed.pos_enc

Traceback of TorchScript, original code (most recent call last):
RuntimeError: "addmm_impl_cpu_" not implemented for 'Half'

  0%|                                                                                                                                                                                          | 0/1 [00:19<?, ?it/s]
Traceback (most recent call last):
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\gradio\queueing.py", line 521, in process_events
    response = await route_utils.call_process_api(
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\gradio\route_utils.py", line 276, in call_process_api
    output = await app.get_blocks().process_api(
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\gradio\blocks.py", line 1945, in process_api
    result = await self.call_function(
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\gradio\blocks.py", line 1525, in call_function
    prediction = await utils.async_iteration(iterator)
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\gradio\utils.py", line 655, in async_iteration
    return await iterator.__anext__()
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\gradio\utils.py", line 648, in __anext__
    return await anyio.to_thread.run_sync(
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\anyio\to_thread.py", line 56, in run_sync
    return await get_async_backend().run_sync_in_worker_thread(
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\anyio\_backends\_asyncio.py", line 2364, in run_sync_in_worker_thread
    return await future
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\anyio\_backends\_asyncio.py", line 864, in run
    result = context.run(func, *args)
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\gradio\utils.py", line 631, in run_sync_iterator_async
    return next(iterator)
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\gradio\utils.py", line 814, in gen_wrapper
    response = next(iterator)
  File "webui.py", line 120, in generate_audio
    for i in cosyvoice.inference_zero_shot(tts_text, prompt_text, prompt_speech_16k, stream=stream, speed=speed):
  File "D:\projects\ai\CosyVoice\cosyvoice\cli\cosyvoice.py", line 73, in inference_zero_shot
    for model_output in self.model.tts(**model_input, stream=stream, speed=speed):
  File "D:\projects\ai\CosyVoice\cosyvoice\cli\model.py", line 191, in tts
    this_tts_speech = self.token2wav(token=this_tts_speech_token,
  File "D:\projects\ai\CosyVoice\cosyvoice\cli\model.py", line 104, in token2wav
    tts_mel, flow_cache = self.flow.inference(token=token.to(self.device),
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\torch\utils\_contextlib.py", line 115, in decorate_context
    return func(*args, **kwargs)
  File "D:\projects\ai\CosyVoice\cosyvoice\flow\flow.py", line 123, in inference
    token = self.input_embedding(torch.clamp(token, min=0)) * mask
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\torch\nn\modules\module.py", line 1501, in _call_impl
    return forward_call(*args, **kwargs)
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\torch\nn\modules\sparse.py", line 162, in forward
    return F.embedding(
  File "D:\ProgramData\miniconda3\envs\cosyvoice\lib\site-packages\torch\nn\functional.py", line 2210, in embedding
    return torch.embedding(weight, input, padding_idx, scale_grad_by_freq, sparse)
RuntimeError: Expected tensor for argument #1 'indices' to have one of the following scalar types: Long, Int; but got torch.FloatTensor instead (while checking arguments for embedding)

[keyiis]

感谢钉钉群@铝箱-CosyVoice开发者的解答，在webui.py的第184行把初始化代码改为下面这样就可以运行了。 cosyvoice = CosyVoice(args.model_dir,load_jit=False, fp16=False) 但是速度奇慢，点击生成音频用了300秒，我的电脑配置也不算很低，有什么办法改进速度吗？

显卡是3050 4gb显存

[aluminumbox]

感谢钉钉群@铝箱-CosyVoice开发者的解答，在webui.py的第184行把初始化代码改为下面这样就可以运行了。 cosyvoice = CosyVoice(args.model_dir,load_jit=False, fp16=False) 但是速度奇慢，点击生成音频用了300秒，我的电脑配置也不算很低，有什么办法改进速度吗？ 显卡是3050 4gb显存

应该是你gpu没用上，看报错是你这边不支持half