This week, we released some improvements in the timestamp annotation endpoint and a bugfix in TTS:
- annotation is now possible for all languages supported by Azure (most of our voices);
- STT works better as it’s not stopping on short silences anymore;
- multi-section annotation is much faster;
- fixes an unrelated, but influencing bug where cached speech sections would have incorrectly saved names;
- API reference for TTS/annotate has been updated.
- We added a guide to using the files and folders management endpoints in the AudioStack API.
- We added a usage guide for generating a script using AI to the docs.
- Clarified some wording to make the terms more understandable.