Diarization paramter updates (prefer current speaker) #120
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
chrisw77
force-pushed
the
Feature/speaker_diarization_prefer_current_speaker
branch
2 times, most recently
from
38ce8c3 to
7f94d0a
Compare
…posed [speaker_sensitivity] for RT (was batch only).
chrisw77
force-pushed
the
Feature/speaker_diarization_prefer_current_speaker
branch
from
7f94d0a to
e62b963
Compare
|
LGTM |
dumitrugutu
approved these changes
Apr 18, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Added new parameter prefer_current_speaker for Diarization. This boolean parameter defaults to false, but when enabled instructs the clustering algorithm to ask an extra question "Am I close enough to current speaker?" before going on to the usual "Which speaker is closest?". Thus, if the current word level speaker embedding is similar enough to the current active speaker (i.e. that which was output for the previous word) it will stay with that speaker, even if another speaker is now closer. This is to reduce occasional situations where we can flip-flop back and forth between similar speakers even when only one speaker is speaking (typically on long, noisy audio)
Also exposed speaker_sensitivity for RT (was batch only).