wx44wx/pipecat
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update NVIDIA STT services for Nemotron Speech defaults and config parity (#4269)

Browse Source 
* Update NVIDIA STT services for Nemotron Speech defaults and config parity

* Add changelog entry for PR #4269

* initialize boosted LM settings defaults in streaming STT

* Align NVIDIA STT language handling with other STT services

* add finalised flag to Nvidia stt final transcripts, remove processing latency logs

* Changing interim transcription logging to tracing.

---------

Co-authored-by: sathwika <geereddysath@nvidia.com>
Co-authored-by: filipi87 <filipi87@gmail.com>
This commit is contained in:
Sathwika Reddy Geereddy
2026-04-23 18:31:27 +05:30
committed by GitHub
parent 4d14251f4a
commit 21f6c2afa5
2 changed files with 212 additions and 111 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
changelog/4269.changed.md Normal file
View File

@@ -0,0 +1,2 @@
- Updated NVIDIA STT services to align with Nemotron Speech defaults and configuration: `api_key` is now optional for local deployments, additional recognition settings are available (including alternatives, word offsets, and diarization), and streaming/segmented docs now reflect Nemotron Speech APIs.
- NVIDIA streaming STT now sets `TranscriptionFrame.finalized=True` when the provider marks a result as final, and preserves `language` on both `TranscriptionFrame` and `InterimTranscriptionFrame`.