Skip to content

Conversation

@vimalmanohar
Copy link
Owner

@vimalmanohar vimalmanohar commented Nov 24, 2016

This pull request is a place holder for all the modifications required for training nnet3 models for speech activity detection, music detection and similar tasks and using those models on test data. This includes recipes for

  1. Training SAD and music detection model on Fisher and Babel corpus corrupted with MUSAN music and noise.
  2. Creating segments for Babel dev data (evalutated on WER), AMI dev data (evalutated on DER), RT04 (evaluated on DER)
  3. Music detection task on Broadcast news

This PR request also contains generic tools for segmentation, nnet3 modifications for training with multi-task objectives and other generic utility scripts and binaries.

@vimalmanohar vimalmanohar force-pushed the asr_diarization_clean branch from 0fde60d to 46ced5d Compare December 3, 2016 02:03
@vimalmanohar vimalmanohar force-pushed the asr_diarization_clean branch from 46ced5d to 398ece6 Compare December 3, 2016 04:05
@vimalmanohar
Copy link
Owner Author

Creating new PR wrt master #4

vimalmanohar pushed a commit that referenced this pull request Jan 13, 2019
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

Successfully merging this pull request may close these issues.

4 participants