Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

xprobe/xinference:v1.1.1-cpu中无法使用fishspeech1.5 #2739

Open
1 of 3 tasks
zhudemiao opened this issue Jan 5, 2025 · 4 comments
Open
1 of 3 tasks

xprobe/xinference:v1.1.1-cpu中无法使用fishspeech1.5 #2739

zhudemiao opened this issue Jan 5, 2025 · 4 comments
Milestone

Comments

@zhudemiao
Copy link

System Info / 系統信息

xprobe/xinference:v1.1.1-cpu

Running Xinference with Docker? / 是否使用 Docker 运行 Xinfernece?

  • docker / docker
  • pip install / 通过 pip install 安装
  • installation from source / 从源码安装

Version info / 版本信息

xprobe/xinference:v1.1.1-cpu

The command used to start Xinference / 用以启动 xinference 的命令

2025-01-05 04:14:25,357 xinference.model.audio.fish_speech 380 WARNING Fish speech does not support setting voice: .
2025-01-05 04:14:25,357 xinference.core.model 380 DEBUG [request 8c80a07a-cb1b-11ef-95c7-0242c0a8b002] Leave speech, elapsed time: 0 s
2025-01-05 04:14:25,357 xinference.core.model 380 DEBUG After request speech, current serve request count: 0 for the model FishSpeech-1.5-0
2025-01-05 04:14:25,372 xinference.core.model 380 ERROR stream encountered an error.
Traceback (most recent call last):
File "/opt/conda/lib/python3.11/site-packages/xinference/core/model.py", line 544, in _handle_pending_requests
r = await asyncio.to_thread(_wrapper, gen)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/lib/python3.11/asyncio/threads.py", line 25, in to_thread
return await loop.run_in_executor(None, func_call)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/lib/python3.11/concurrent/futures/thread.py", line 58, in run
result = self.fn(*self.args, **self.kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/lib/python3.11/site-packages/xinference/core/model.py", line 537, in _wrapper
return next(_gen)
^^^^^^^^^^
File "/opt/conda/lib/python3.11/site-packages/xinference/core/model.py", line 448, in _to_generator
for v in gen:
File "/opt/conda/lib/python3.11/site-packages/xinference/model/audio/fish_speech.py", line 158, in _stream_generator
writer = torchaudio.io.StreamWriter(out, format=response_format)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/lib/python3.11/site-packages/torio/io/_streaming_media_encoder.py", line 197, in init
self._s = ffmpeg_ext.StreamingMediaEncoderFileObj(dst, format, buffer_size)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/lib/python3.11/site-packages/torio/_extension/utils.py", line 25, in getattr
self._import_once()
File "/opt/conda/lib/python3.11/site-packages/torio/_extension/utils.py", line 39, in _import_once
self.module = self.import_func()
^^^^^^^^^^^^^^^^^^
File "/opt/conda/lib/python3.11/site-packages/torio/_extension/utils.py", line 143, in _init_ffmpeg
ext = _find_ffmpeg_extension(ffmpeg_vers)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/lib/python3.11/site-packages/torio/_extension/utils.py", line 122, in _find_ffmpeg_extension
raise ImportError(
ImportError: Failed to intialize FFmpeg extension. Tried versions: ['6', '5', '4', '']. Enable DEBUG logging to see more details about the error.

Reproduction / 复现过程

1、在使用大模型fishspeech1.5时,报告没有找到ffmpeg_extension的错误

Expected behavior / 期待表现

能够正常使用

@XprobeBot XprobeBot added this to the v1.x milestone Jan 5, 2025
@zhudemiao
Copy link
Author

使用xprobe/xinference:v1.1.0-cpu,用fishspeech-1.4时,报告同样的错误。

@qinxuye
Copy link
Contributor

qinxuye commented Jan 6, 2025

容器内 pip show ffmpeg 有结果吗

@zhudemiao
Copy link
Author

(base) root@034f12202d08:/# apt list --installed|grep ffmpeg

WARNING: apt does not have a stable CLI interface. Use with caution in scripts.

ffmpeg/oldstable-security,now 7:4.3.8-0+deb11u1 amd64 [installed]
但是:
容器内 pip show ffmpeg 没有结果,需要安装吗?

@zhudemiao
Copy link
Author

zhudemiao commented Jan 7, 2025

安装之后,报告同样的错误:
pip install --no-cache-dir ffmpeg -i https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple

Looking in indexes: https://mirrors.tuna.tsinghua.edu.cn/pypi/web/simple
Collecting ffmpeg
Downloading https://mirrors.tuna.tsinghua.edu.cn/pypi/web/packages/f0/cc/3b7408b8ecf7c1d20ad480c3eaed7619857bf1054b690226e906fdf14258/ffmpeg-1.4.tar.gz (5.1 kB)
Preparing metadata (setup.py): started
Preparing metadata (setup.py): finished with status 'done'
Building wheels for collected packages: ffmpeg
Building wheel for ffmpeg (setup.py): started
Building wheel for ffmpeg (setup.py): finished with status 'done'
Created wheel for ffmpeg: filename=ffmpeg-1.4-py3-none-any.whl size=6082 sha256=311de2dbe9f9a5a073ba1743849b845c0f3453c32674e35494a25097cdb14a05
Stored in directory: /tmp/pip-ephem-wheel-cache-4foil8u8/wheels/25/07/1a/5817ee6e118eb0cd1a018cdaddea1e11fa2c2bf584494610b6
Successfully built ffmpeg
Installing collected packages: ffmpeg
Successfully installed ffmpeg-1.4

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants