MOSS-Transcribe-Diarize: Transcription and Diarization in One 0.9B Model
vanstriendaniel · x · 2026-08-10
The open-source project MOSS-Transcribe-Diarize has been released, combining transcription and speaker diarization into a single 0.9B parameter model.
- Long Context: Features a 128k context window, processing up to 90 minutes of audio in a single pass without chunking.
- Multilingual: Supports 50+ languages, providing timestamps and speaker labels simultaneously.
- Ready to Use: Eliminates the need to wire up a Whisper-then-pyannote pipeline. It ships with a local subtitle web app where users can upload a video and export SRT files or burn them into an MP4. Licensed under Apache-2.0.
More from coding & agent
- Open Source PaaS Coolify Hits 40k Stars, Offers Free Self-Hosted Vercel Alternative — thisdudelikesAI · 2026-08-10
- Opus 5 Better Than 4.8, But Cutting 80% Instructions Failed: 3 CLAUDE.md Rules Fixed It — PawelHuryn · 2026-08-10
- Open-source T3 Code wins over devs: a unified control plane for Claude Code, Codex, and more — intellectronica · 2026-08-10
- Fluent: Open-Source Kit Turns Claude Code into an Adaptive AI Language Tutor — tom_doerr · 2026-08-10
- Dev Hooks Up Vercel Eve Agent for Automated Troubleshooting and Self-Recovery — chongdashu · 2026-08-10
- RAG-art: Open-Source Local Tool Turns Art History into Structured Prompts — lololerigolo60 · 2026-08-10