The OpenMOSS-Team has released MOSS-Transcribe-Diarize 0.9B, an end-to-end model designed for comprehensive audio understanding. This model performs transcription, speaker diarization, and timestamp generation in a single pass, producing structured, speaker-aware transcripts for long-form audio and video content. It is particularly useful for applications like meetings, calls, and lectures, and can also identify acoustic events. AI
IMPACT This model offers a streamlined approach to audio transcription and speaker diarization, potentially improving efficiency for content analysis and meeting summarization tools.
RANK_REASON Release of a new model with technical details and usage instructions.
- GitHub
- Google Colab
- Hugging Face
- Kaggle
- MOSS-Transcribe-Diarize 0.9B
- OpenMOSS-Team
- OpenMOSS-Team/MOSS-Transcribe-Diarize
- PyTorch
- transformers
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →