Cohere Launches An Open-Source Voice Model Specifically For Transcription

Cohere Launches An Open-Source Voice Model Specifically For Transcription

Cohere Transcribe: The Open Source Voice AI Beating the Big Players Cohere just released its first-ever voice model, and it is making serious noise in the AI space. Called Transcribe, this open source automatic speech recognition model is built for real-world tasks like note-taking, dictation, and speech analysis. With a lean 2-billion-parameter footprint and free API access, it is designed for businesses and developers who want powerful transcription without the heavy price tag. What Is Cohere Transcribe and Why Should You Care? Enterprise AI company Cohere launched Transcribe on March 26, 2026, marking its first major move into the voice AI space. The model is an automatic speech recognition system, meaning it converts spoken audio into accurate written text. What makes it stand out is not just its performance but its accessibility. Unlike many speech models that demand expensive infrastructure, Transcribe is lightweight enough to run on consumer-grade GPUs. That means smaller teams, in…