Dependency errors while running the evaluation notebook for the speech separation uploaded recently

[BNarayanaReddy]

Tested versions

There is a small dependancy error that making unable to instatiate the speech separation pipeline using the command. from pyannote.audio import Pipeline pipeline = Pipeline.from_pretrained( "pyannote/speech-separation-ami-1.0", use_auth_token=True) The corrected dependency is using pyannote-audio version 3.3 instead of 3.1.1 bcoz, in colab it is installing the pyannote-audio 3.1.1 instead. Solution : !pip install -qq speechbrain==0.5.16 !pip install -qq ipython==7.34.0 !pip install -qq ipywidgets openai-whisper whisperx==3.1.5 meeteval !pip install -qq pyannote.audio[separation]==3.3.0

System information

Google Colab - pyannote-audio 3.1.1

Issue description

Bug, while loading the speech separation pre-trained model

# use the following instead
!pip install -qq pyannote.audio[separation]==3.3.0

Minimal reproduction example (MRE)

https://colab.research.google.com/drive/1aheD77AeF7Eq5jmTCrA5ODEBkOolrWAz?usp=sharing

[hbredin]

@clement-pages, can you please have a look and provide a fix?

[BNarayanaReddy]

Solution : !pip install -qq pyannote.audio[separation]==3.3.0

[hbredin]

Thanks @BNarayanaReddy. Would you mind opening a PR?

[BNarayanaReddy]

yes sure. Update pyannote-audio installation version number #1768

[BNarayanaReddy]

Solved the issue for those who run in google colab.