Skip to content

New version of the audio module#123

Merged
sergioburdisso merged 109 commits intoidiap:mainfrom
qanastek:main
Mar 13, 2026
Merged

New version of the audio module#123
sergioburdisso merged 109 commits intoidiap:mainfrom
qanastek:main

Conversation

@qanastek
Copy link
Contributor

No description provided.

qanastek and others added 30 commits November 6, 2025 18:09
- Add annotations saving
- Allow the users to access to the list of requirements
- Completly change the way parameters are given
- Update the tutorials 18 with the new features
- Test all new features
…dio, in order to sort them by modalities.

- Update the documentation.
- Run some tests using the tutorial.
- Delete annotator.py file since unused now.
- Add multi-dialog task annotation
- Update the tutorial
…e work

- Update tutorials for the new parameters of SDialog.audio pipeline
- Split the tutorial in two
- Add results2result method
- Reverse the order or evaluate (dialogs and evaluators)
- Add a plotting method
- Add multiple plots choices
- Update tutorial 8
…ctly parsed and step 3 is now an object.

- Fix impossible to get audio data from turn when loaded from local serialized turn. Now it load the file from its corresponding path if the private audio data is empty.
- Make the color class serializable
- Add multiple ways to display results from an evaluation (raw object, table or structured text)
- Merge differents plots and data together when printing figures about metrics
- Add a maximum number of turn to process for each dialogs in order to reduce CPU load
- Cast all the restults into the same type (float)
- Add the computation of EER and ROC based on the x-vectors embeddings from pyannote
- Update the tutorials and tests
@codecov
Copy link

codecov bot commented Mar 12, 2026

Codecov Report

❌ Patch coverage is 32.26891% with 403 lines in your changes missing coverage. Please review.
✅ Project coverage is 50.76%. Comparing base (6d23ea4) to head (cc18caf).
⚠️ Report is 111 commits behind head on main.

Files with missing lines Patch % Lines
src/sdialog/audio/dialog.py 19.90% 165 Missing ⚠️
src/sdialog/audio/tts/qwen3/tts.py 4.76% 60 Missing ⚠️
src/sdialog/audio/pipeline.py 18.33% 49 Missing ⚠️
src/sdialog/audio/dscaper_utils.py 37.70% 38 Missing ⚠️
src/sdialog/audio/normalizers.py 43.85% 32 Missing ⚠️
src/sdialog/audio/__init__.py 20.00% 12 Missing ⚠️
src/sdialog/audio/room.py 63.33% 11 Missing ⚠️
src/sdialog/audio/tts/index/tts.py 10.00% 9 Missing ⚠️
src/sdialog/audio/tts/huggingface/tts.py 12.50% 7 Missing ⚠️
src/sdialog/audio/tts/kokoro/tts.py 12.50% 7 Missing ⚠️
... and 4 more
Additional details and impacted files
@@            Coverage Diff             @@
##             main     #123      +/-   ##
==========================================
- Coverage   53.00%   50.76%   -2.24%     
==========================================
  Files          43       44       +1     
  Lines        6928     7365     +437     
==========================================
+ Hits         3672     3739      +67     
- Misses       3256     3626     +370     
Files with missing lines Coverage Δ
src/sdialog/__init__.py 62.70% <ø> (ø)
src/sdialog/audio/tts/__init__.py 100.00% <100.00%> (ø)
src/sdialog/audio/tts/base.py 84.61% <100.00%> (ø)
src/sdialog/audio/turn.py 96.55% <85.71%> (-3.45%) ⬇️
src/sdialog/audio/voice_database.py 86.47% <0.00%> (-0.55%) ⬇️
src/sdialog/audio/acoustics_simulator.py 73.86% <44.44%> (-4.19%) ⬇️
src/sdialog/audio/utils.py 76.58% <91.22%> (-1.71%) ⬇️
src/sdialog/audio/tts/huggingface/tts.py 30.43% <12.50%> (-9.57%) ⬇️
src/sdialog/audio/tts/kokoro/tts.py 23.07% <12.50%> (-4.71%) ⬇️
src/sdialog/audio/tts/index/tts.py 35.00% <10.00%> (-11.16%) ⬇️
... and 7 more
🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

@sergioburdisso sergioburdisso merged commit dd3e674 into idiap:main Mar 13, 2026
4 checks passed
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.

3 participants