Back to News
Cohere•March 26, 2026
Cohere Transcribe
Open Source
open-source
speech recognition
automatic speech recognition
enterprise AI
### TL;DR
Transcribe is Cohere's open-source automatic speech recognition (ASR) model designed to convert audio data into precise text for search, analytics, and automation. Supporting 14 languages, it offers high accuracy and efficiency for enterprise applications.
Key Insights & Metrics
Pricing
Free via API or Download via Hugging Face
Cost structure
Version
03-2026
Current release version
Hardware
Compatible modern consumer-grade or cloud GPU for self-hosting (minimum 8 GB VRAM)
Compute requirements
Category
Open Source
Licensing model
Region
Canada
Primary region
Key Features
- Supports 14 languages: English, French, German, Italian, Spanish, Portuguese, Greek, Dutch, Polish, Vietnamese, Chinese, Arabic, Japanese, Korean
- High accuracy with a word error rate (WER) of 5.42 on the Hugging Face Open ASR leaderboard
- Real-time processing capability, transcribing 525 minutes of audio per minute
- Open-source model with Apache 2.0 license for flexible integration
- Available through Cohere's API for free, with options for private deployment via Model Vault
Discussion
0
Upvotes
0
Downvotes
0 reviews
Sign in to leave a review
Reviews
No reviews yet. Be the first to review!