Skip to content

Add JSON, SRT, and WebVTT ASR exports - #128

Open
jmayank1511 wants to merge 1 commit into
nvidia-riva:mainfrom
jmayank1511:feature/asr-transcript-export
Open

Add JSON, SRT, and WebVTT ASR exports#128
jmayank1511 wants to merge 1 commit into
nvidia-riva:mainfrom
jmayank1511:feature/asr-transcript-export

Conversation

@jmayank1511

Copy link
Copy Markdown
Contributor

Summary

  • add shared structured JSON, SRT, and WebVTT export support
  • support both riva_asr_client and riva_streaming_asr_client, with extension inference or explicit --output_format
  • automatically request word timestamps for subtitle output, preserve legacy JSON text, and report invalid or failed exports
  • keep file export independent from console transcript printing

Testing

  • bazel test --workspace_status_command=/bin/true //riva/clients/asr:transcript_exporter_test
  • built //riva/clients/asr:riva_asr_client and //riva/clients/asr:riva_streaming_asr_client
  • live 30-second and 5-minute JSON/SRT/VTT matrices passed against public Parakeet TDT offline and Parakeet CTC streaming NIMs
  • validated JSON structure, subtitle syntax, cue ordering, word timestamps, and cross-format transcript consistency

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