Feature Request: Add NVIDIA Nemotron 3.5 Transcribe Support via NeMo-Speech.cpp

Open
  • Is something wrong?
#16,524 1 comment 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
5/5
Estimated time
Over a week
Newbie friendliness
35/100
Issue type
Feature
Clarity
Mostly clear
Activity status
Active
Tech stack
cpp, csharp
Domain
audio-video-rtc, desktop

Research direction

Start by reading Subtitle Edit’s existing OpenAI Compatible Server speech-to-text integration, then compare its transcription API expectations with the NeMo-Speech.cpp project linked in the issue. Decide which integration approach fits and verify how word-level timestamps and speaker information can be imported. Done means Nemotron 3.5 transcriptions work in Speech-to-Text while preserving that information.

Written by the indexing model from the issue text.

Description

Hello,

I would like to request support for NVIDIA’s new Nemotron 3.5 Transcribe model in Subtitle Edit’s Speech-to-Text functionality.

NVIDIA has released NeMo-Speech.cpp, an open-source C++ inference runtime that supports Nemotron 3.5 Transcribe and provides an OpenAI-compatible HTTP API.

Project:
https://github.com/NVIDIA/NeMo-Speech.cpp

The model offers several interesting capabilities:

  • Word-level timestamps
  • Speaker diarization
  • Multilingual transcription (40 languages)
  • Speech translation
  • Context-aware transcription
  • Recordings up to one hour long

Possible integration approaches:

  1. Add NeMo-Speech.cpp as a dedicated Speech-to-Text engine.
  2. Improve the existing OpenAI Compatible Server integration to support NeMo-Speech.cpp’s transcription API and its word-level timestamp output.

Ideally, Subtitle Edit should preserve word-level timestamps and speaker information when importing the transcription.

This would provide users with another powerful local transcription option, without requiring cloud services.

Thank you for considering this feature!

Dominant language
C#
Stars
14.8k
Forks
1.4k
Avg merge
58m
Merged PRs (30d)
697

Getting set up

This project ships no dev container, Dockerfile or contributing guide, so setting up is up to you: start from its README, and see our first-contribution guide for the general steps.

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from SubtitleEdit/subtitleedit

All issues in SubtitleEdit/subtitleedit

Similar issues

More C# issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.