⚙️ AI Hardware

Cohere's 2B Transcribe Model Crushes ASR Benchmarks—But Only in English?

What if top-tier speech-to-text ran on your gaming rig, not a data center? Cohere's Transcribe just made that real, smashing rivals on benchmarks while staying open source.

James Kowalski 📅 Mar 29, 2026 ⏱️ 3 min read 👁️ 3 views
Cohere Transcribe model dominating Hugging Face Open ASR leaderboard graph

⚡ Key Takeaways

  • Transcribe tops Hugging Face ASR leaderboard at 5.42% WER, beating Zoom, IBM, ElevenLabs.
  • 2B params enable consumer GPU self-hosting; processes 525x realtime audio.
  • Enterprise play via free API and North integration; multilingual gaps in key languages.

🧠 What's your take on this?

Cast your vote and see what theAIcatchup readers think

James Kowalski
Written by

James Kowalski

Investigative tech reporter focused on AI ethics, regulation, and societal impact.

#ASR model #Cohere #Voice AI #open source ASR #open source transcription #speech recognition #speech transcription

Worth sharing?

Get the best AI stories of the week in your inbox — no noise, no spam.

Originally reported by TechCrunch - AI

Related Stories

📬

Stay in the loop

The week's most important stories from theAIcatchup, delivered once a week.

No spam. Unsubscribe any time.