Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Sortformer Diarizer 4spk v1 model PR Part 4: Sortformer Documents and Notebook Tutorials #11707

Draft
wants to merge 4 commits into
base: main
Choose a base branch
from
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
Expand Up @@ -5,9 +5,9 @@
# These parameters were optimized on CallHome Dataset from the NIST SRE 2000 Disc8, especially from the part1 (callhome1) specified in: Kaldi, “Kaldi x-vector recipe v2,” https://github.com/kaldi-asr/kaldi/blob/master/egs/callhome_diarization/v2/run.sh
# Trial 24682 finished with value: 0.10257785779242055 and parameters: {'onset': 0.53, 'offset': 0.49, 'pad_onset': 0.23, 'pad_offset': 0.01, 'min_duration_on': 0.42, 'min_duration_off': 0.34}. Best is trial 24682 with value: 0.10257785779242055.
parameters:
onset: 0.53 # Onset threshold for detecting the beginning and end of a speech
offset: 0.49 # Offset threshold for detecting the end of a speech
pad_onset: 0.23 # Adding durations before each speech segment
pad_offset: 0.01 # Adding durations after each speech segment
min_duration_on: 0.42 # Threshold for small non-speech deletion
min_duration_off: 0.34 # Threshold for short speech segment deletion
onset: 0.53 # Onset threshold for detecting the beginning of a speech segment
offset: 0.49 # Offset threshold for detecting the end of a speech segment
pad_onset: 0.23 # Adds the specified duration at the beginning of each speech segment
pad_offset: 0.01 # Adds the specified duration at the end of each speech segment
min_duration_on: 0.42 # Removes short silences if the duration is less than the specified minimum duration
min_duration_off: 0.34 # Removes short speech segments if the duration is less than the specified minimum duration
Original file line number Diff line number Diff line change
Expand Up @@ -5,9 +5,9 @@
# These parameters were optimized on the development split of DIHARD3 dataset (See https://arxiv.org/pdf/2012.01477).
# Trial 732 finished with value: 0.12171946949255649 and parameters: {'onset': 0.64, 'offset': 0.74, 'pad_onset': 0.06, 'pad_offset': 0.0, 'min_duration_on': 0.1, 'min_duration_off': 0.15}. Best is trial 732 with value: 0.12171946949255649.
parameters:
onset: 0.64 # Onset threshold for detecting the beginning and end of a speech
offset: 0.74 # Offset threshold for detecting the end of a speech
pad_onset: 0.06 # Adding durations before each speech segment
pad_offset: 0.0 # Adding durations after each speech segment
min_duration_on: 0.1 # Threshold for small non-speech deletion
min_duration_off: 0.15 # Threshold for short speech segment deletion
onset: 0.64 # Onset threshold for detecting the beginning of a speech segment
offset: 0.74 # Offset threshold for detecting the end of a speech segment
pad_onset: 0.06 # Adds the specified duration at the beginning of each speech segment
pad_offset: 0.0 # Adds the specified duration at the end of each speech segment
min_duration_on: 0.1 # Removes short silences if the duration is less than the specified minimum duration
min_duration_off: 0.15 # Removes short speech segments if the duration is less than the specified minimum duration
Loading
Loading