GameCult/AquaSynth
Watch
9
0
Fork
You've already forked AquaSynth
 0
Code Issues Pull requests 4 Projects Releases Packages Wiki Activity Actions

Add log-mel evidence to eSpeak workout

Browse source
This commit is contained in:
Meta Vens 2026-05-19 16:15:44 +01:00
commit fdb9d76cfd
3 changed files with 66 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

7
docs/ipa-vocal-tract-roadmap.md
View file

@ -455,9 +455,10 @@ artifacts/parity/espeak-ng-ipa-workout/<timestamp>-tiny/
```
The report records the source text, eSpeak IPA transcription, sample counts, and
basic audio features. This is the first external speech judge for the vocal
tract lane. It should pressure AquaSynth's future IPA/gesture/render path, not
replace it.
basic audio features. It also records normalized log-mel spectrogram summaries
per fixture and a compact log-mel distance matrix across the workout. This is
the first external speech judge for the vocal tract lane. It should pressure
AquaSynth's future IPA/gesture/render path, not replace it.
## First Work Packets