get clustering labels #4

Activity

1. mzboito commented on Oct 24, 2024

   

   mzboito

   on Oct 24, 2024

   Contributor

   More actions

   Hello,

   Many thanks for the interest in our work!

   Please be aware that the available k-means model (https://huggingface.co/utter-project/mHuBERT-147/blob/main/mhubert147_faiss.index) was trained on the 2nd iteration model (available here: https://huggingface.co/utter-project/mHuBERT-147-base-2nd-iter).
   It was NOT trained on the checkpoint_best.pt from the 3rd iteration.

   If you want to generate discrete labels using features from the 3rd iteration model (https://huggingface.co/utter-project/mHuBERT-147/blob/main/checkpoint_best.pt) you will need to train a new k-means model.

   The procedure for doing so is the following:

   1. Create manifest files for your data and extract features using this script: https://github.com/utter-project/mHuBERT-147-scripts/blob/main/02_feature_extraction/hubert_feature_extraction.sh (this will produce .npy and .len features)
   2. Train the model using your extracted features: https://github.com/utter-project/mHuBERT-147-scripts/tree/main/03_faiss_indices
   3. Extract the labels using the script you mentioned (https://github.com/utter-project/mHuBERT-147-scripts/blob/main/03_faiss_indices/apply_index_per_file.py)
2. taalua commented on Oct 24, 2024

   

   taalua

   on Oct 24, 2024

   Author

   More actions

   Thank you for your prompt response. Just to clarify, If I want to get discrete label from 2nd iteration, I can use this script https://github.com/utter-project/mHuBERT-147-scripts/blob/main/03_faiss_indices/apply_index_per_file.py

   Thank you.
3. mzboito commented on Oct 25, 2024

   

   mzboito

   on Oct 25, 2024

   Contributor

   More actions

   Yes. You just need to extract the features first!