se_extractor.get_se递归调用过深

[keqixiaojie]

Traceback (most recent call last):
File "/root/shared-nvme/OpenVoice/part3.py", line 17, in
target_se, audio_name = se_extractor.get_se(reference_speaker, tone_color_converter, vad=False)
File "/root/shared-nvme/OpenVoice/openvoice/se_extractor.py", line 152, in get_se
return vc_model.extract_se(audio_segs, se_save_path=se_path), audio_name
File "/root/shared-nvme/OpenVoice/openvoice/api.py", line 123, in extract_se
audio_ref, sr = librosa.load(fname, sr=hps.data.sampling_rate)
File "/root/shared-nvme/conda/envs/openvoice/lib/python3.9/site-packages/librosa/util/decorators.py", line 88, in inner_f
return f(*args, **kwargs)
File "/root/shared-nvme/conda/envs/openvoice/lib/python3.9/site-packages/librosa/core/audio.py", line 183, in load
y = resample(y, orig_sr=sr_native, target_sr=sr, res_type=res_type)
File "/root/shared-nvme/conda/envs/openvoice/lib/python3.9/site-packages/librosa/util/decorators.py", line 88, in inner_f
return f(*args, **kwargs)
File "/root/shared-nvme/conda/envs/openvoice/lib/python3.9/site-packages/librosa/core/audio.py", line 617, in resample
y_hat = resampy.resample(y, orig_sr, target_sr, filter=res_type, axis=-1)
File "/root/shared-nvme/conda/envs/openvoice/lib/python3.9/site-packages/resampy/core.py", line 168, in resample
resample_f_s(
File "/root/shared-nvme/conda/envs/openvoice/lib/python3.9/site-packages/numba/np/ufunc/gufunc.py", line 279, in call
return self.ufunc(*args, **kwargs)
RecursionError: maximum recursion depth exceeded while calling a Python object
请问这该怎么解决

[dosubot]

@keqixiaojie 你好！我是Dosu，我是一个可以帮助你解决 bug、回答问题并成为贡献者的机器人。我会在等待人类维护者的时候帮助你。

The RecursionError: maximum recursion depth exceeded while calling a Python object typically occurs when a function calls itself too many times without a base case to stop the recursion. In the context of the resample function from librosa, this error might be related to the resampling process, possibly due to an issue with the input data or the parameters being used.

Here are a few steps you can take to troubleshoot and potentially resolve this issue:

1. Check Input Audio Files: Ensure that the audio files being processed are not corrupted and are in a supported format. Sometimes, issues with the audio file itself can cause unexpected behavior.

2. Sampling Rate Mismatch: Verify that the original sampling rate of the audio files matches the expected sampling rate (hps.data.sampling_rate). If there's a mismatch, it might cause excessive resampling operations.

3. Update Libraries: Make sure that you are using the latest versions of librosa and resampy. There might be bug fixes or improvements in newer versions that address this issue.

4. Modify Recursion Limit: As a temporary workaround, you can increase the recursion limit in Python using sys.setrecursionlimit(new_limit). However, this is not a recommended long-term solution as it might lead to other issues.

5. Alternative Resampling Methods: If the problem persists, consider using a different resampling method or library to handle the audio resampling process.

If these steps do not resolve the issue, you may need to further investigate the specific conditions under which the error occurs, such as the size and content of the audio files or the specific parameters used in the resample function.

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other}

[keqixiaojie]

解决了，重装了librosa