[Bug] AttributeError: 'TTS' object has no attribute 'is_multi_lingual'

[Niggling]

Describe the bug

因为huggingface.co连接有问题所以下载的本地模型，最后出现了奇怪的错误
The local model downloaded because there is a problem with the connection in huggingface.co, and finally there is a strange error.

To Reproduce

import torch
from TTS.api import TTS

Get device

device = "cuda" if torch.cuda.is_available() else "cpu"

Init TTS with local model path

model_path = "C:/Code/Coqui/TTS/model/tts_models--zh-CN--baker--tacotron2-DDC-GST" # Update this path
tts = TTS(model_path=model_path + "/model_file.pth", config_path=model_path + "/config.json").to(device)

Run TTS

❗ Since this model is multi-lingual voice cloning model, we must set the target speaker_wav and language

Text to speech list of amplitude values as output

wav = tts.tts(text="你好世界!", speaker_wav="my/cloning/audio.wav", language="zh-cn")

Text to speech to a file

tts.tts_to_file(text="你好世界!", speaker_wav="my/cloning/audio.wav", language="zh-cn", file_path="output.wav")

model/tts_models--zh-CN--baker--tacotron2-DDC-GST中只有config.json model_file.pth scale_stats.npy

Expected behavior

No response

Logs

C:\Users\ning\AppData\Local\Programs\Python\Python39\python.exe C:/Code/Coqui/TTS/run.py
C:\Code\Coqui\TTS\TTS\utils\audio\processor.py:6: UserWarning: A NumPy version >=1.22.4 and <1.29.0 is required for this version of SciPy (detected version 1.22.0)
  import scipy.io.wavfile
 > Using model: tacotron2
 > Setting up Audio Processor...
 | > sample_rate:22050
 | > resample:False
 | > num_mels:80
 | > log_func:np.log10
 | > min_level_db:-100
 | > frame_shift_ms:None
 | > frame_length_ms:None
 | > ref_level_db:0
 | > fft_size:1024
 | > power:1.5
 | > preemphasis:0.0
 | > griffin_lim_iters:60
 | > signal_norm:True
 | > symmetric_norm:True
 | > mel_fmin:50.0
 | > mel_fmax:7600.0
 | > pitch_fmin:0.0
 | > pitch_fmax:640.0
 | > spec_gain:1.0
 | > stft_pad_mode:reflect
 | > max_norm:4.0
 | > clip_norm:True
 | > do_trim_silence:True
 | > trim_db:60
 | > do_sound_norm:False
 | > do_amp_to_db_linear:True
 | > do_amp_to_db_mel:True
 | > do_rms_norm:False
 | > db_level:None
 | > stats_path:C:\Users\ning\AppData\Local\tts\tts_models--zh-CN--baker--tacotron2-DDC-GST\scale_stats.npy
 | > base:10
 | > hop_length:256
 | > win_length:1024
Traceback (most recent call last):
  File "C:\Code\Coqui\TTS\run.py", line 14, in <module>
    wav = tts.tts(text="你好世界!", speaker_wav="my/cloning/audio.wav", language="zh-cn")
  File "C:\Code\Coqui\TTS\TTS\api.py", line 273, in tts
    self._check_arguments(
  File "C:\Code\Coqui\TTS\TTS\api.py", line 228, in _check_arguments
    if self.is_multi_lingual and language is None:
  File "C:\Users\ning\AppData\Local\Programs\Python\Python39\lib\site-packages\torch\nn\modules\module.py", line 1688, in __getattr__
    raise AttributeError(f"'{type(self).__name__}' object has no attribute '{name}'")
AttributeError: 'TTS' object has no attribute 'is_multi_lingual'
 > Model's reduction rate `r` is set to: 2

进程已结束，退出代码为 1

Environment

Win10

{
    "CUDA": {
        "GPU": [],
        "available": false,
        "version": null
    },
    "Packages": {
        "PyTorch_debug": false,
        "PyTorch_version": "2.2.0+cpu",
        "TTS": "0.22.0",
        "numpy": "1.22.0"
    },
    "System": {
        "OS": "Windows",
        "architecture": [
            "64bit",
            "WindowsPE"
        ],
        "processor": "Intel64 Family 6 Model 165 Stepping 2, GenuineIntel",
        "python": "3.9.0",
        "version": "10.0.19041"
    }
}

Additional context

No response

[Niggling]

新发现了这个

[weisiwu]

同样的问题，参考这个人的解决方法
#3449
这是他的修改commit
9108720
修改 api.py 102 行的判断可解决。

[weisiwu]

你也在弄转语音吗，我们可以加一波

[Niggling]

你也在弄转语音吗，我们可以加一波

ok，但我只是用下这个来做毕设

[weisiwu]

你也在弄转语音吗，我们可以加一波

ok，但我只是用下这个来做毕设

qq
2436887475

[eginhard]

It's fixed in our fork: https://github.com/idiap/coqui-ai-TTS

[DeepanshCUHK]

Go to /.venv/lib/python3.9/site-packages/TTS/api.py

inside init function, change self.config = load_config(config_path) if config_path else None to self.config = None

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. You might also look our discussion channels.

[hussain0613]

It's fixed in our fork: https://github.com/idiap/coqui-ai-TTS

Hi @eginhard, it seems this repo does not have that fix yet.
why not create a PR from your fork here?
that would be really helpful.

[eginhard]

@hussain0613 Yes, because this repo is not maintained anymore. You should use the fork, you can install it with pip install coqui-tts.

[hussain0613]

@hussain0613 Yes, because this repo is not maintained anymore. You should use the fork, you can install it with pip install coqui-tts.
Ya, that's unfortunate.
And thank you. started using that.