[Help]: RuntimeError: espeak not installed on your system

[chenwei20]

When I execute python maskgct_inference.py on Windows, the following prompt appears

./models/tts/maskgct/g2p\sources\g2p_chinese_model\poly_bert_model.onnx
E:\Amphion\envs\amphion\lib\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py:69: UserWarning: Specified provider 'CUDAExecutionProvider' is not in available provider names.Available providers: 'AzureExecutionProvider, CPUExecutionProvider'
  warnings.warn(
Traceback (most recent call last):
  File "E:\Amphion\models\tts\maskgct\maskgct_inference.py", line 6, in <module>
    from models.tts.maskgct.maskgct_utils import *
  File "E:\Amphion\models\tts\maskgct\maskgct_utils.py", line 26, in <module>
    from models.tts.maskgct.g2p.g2p_generation import g2p, chn_eng_g2p
  File "E:\Amphion\models\tts\maskgct\g2p\g2p_generation.py", line 10, in <module>
    from models.tts.maskgct.g2p.utils.g2p import phonemizer_g2p
  File "E:\Amphion\models\tts\maskgct\g2p\utils\g2p.py", line 17, in <module>
    phonemizer_zh = EspeakBackend(
  File "E:\Amphion\envs\amphion\lib\site-packages\phonemizer\backend\espeak\espeak.py", line 45, in __init__
    super().__init__(
  File "E:\Amphion\envs\amphion\lib\site-packages\phonemizer\backend\espeak\base.py", line 39, in __init__
    super().__init__(
  File "E:\Amphion\envs\amphion\lib\site-packages\phonemizer\backend\base.py", line 77, in __init__
    raise RuntimeError(  # pragma: nocover
RuntimeError: espeak not installed on your system
(amphion) PS E:\Amphion>

I have added espeak to the Path variable in PowerShell and also added it to the user environment variables, but why does it still say that espeak is not installed?

[blues-green]

The same for Linux. How do I install it?

[HeCheng0625]

Hi, try to "sudo apt-get install espeak-ng"?

[yuantuo666]

Hi, the MaskGCT is built in a Linux environment. For a better coding experience, it is recommended that Linux be used to reproduce.
Besides, we prefer English issues in this repository. (all Chinese question has been translated in to English via Google Translate)
According to the stack trace, it seems to be the problem with the phonemizer package, reading their docs might be helpful.

[yuantuo666]

For people who are having problem in configuring the env on a Windows machine, you can try to follow this blog post: https://www.cnblogs.com/v3ucn/p/18511187

[justinjohn0306]

Windows users can try my fork here: https://github.com/justinjohn0306/MaskGCT-Windows

[ytlviv]

I also encountered the same problem, I am mac os, through brew install espeak installation, also reported this error, how to deal with

[AimForNaN]

I'm getting the same error on my Linux machine. I tried adjusting env vars for both conda and shell session, but nothing seems to work.

[AimForNaN]

Coming across this issue, I placed the following code at the top of gradio_demo.py:

from phonemizer.backend.espeak.wrapper import EspeakWrapper
EspeakWrapper()

Interestingly, this was sufficient to get passed the error. So, it's not so much that it can't find espeak as much as it is that the runtime error message is a lie.