Skip to content

Add AssemblyAI STT 3.7 transcription and realtime models - #443

Merged
ProfSynapse merged 1 commit into
mainfrom
feat/assemblyai-stt-3-7
Nov 18, 2026
Merged

ProfSynapse merged 1 commit into
mainfrom
feat/assemblyai-stt-3-7

Conversation

@ProfSynapse

Copy link
Copy Markdown
Owner

AssemblyAI STT 3.7 is available for recorded audio and realtime sessions, but Nexus still offers the previous generation by default. Add the recorded model and both realtime variants, preferring Realtime Agent for voice chat while retaining saved older selections. Disable fillers on recorded STT 3.7 requests to preserve clean transcripts.

Update the model verification guidance to cover dedicated speech catalogs outside the chat registry checker.

Validation: 106 targeted catalog, adapter, session, and service tests; schema and mobile gates; lint, TypeScript, and production build stages. Two existing unrelated lint warnings remain. Live API verification was unavailable because no AssemblyAI credential was configured in the environment or repository .env.

Source: https://www.assemblyai.com/blog/introducing-assemblyai-stt-3-7

@ProfSynapse
ProfSynapse merged commit d8bf505 into main Nov 18, 2026
1 check passed
@ProfSynapse
ProfSynapse deleted the feat/assemblyai-stt-3-7 branch November 18, 2026 19:29
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.

1 participant