Cohere Transcribe: Speech Recognition
9 hours ago
- #ASR
- #AI
- #Open Source
- Cohere released Transcribe, an open-source automatic speech recognition (ASR) model designed for practical use.
- The model ranks #1 on HuggingFace's Open ASR Leaderboard with a 5.42% average word error rate, outperforming competitors like Whisper Large v3.
- It supports 14 languages, has a 2B parameter architecture, and is licensed under Apache 2.0.
- Transcribe offers high throughput and efficiency, suitable for real-time applications and production environments.
- Available via Hugging Face, Cohere's API, and Model Vault for managed deployment.