<p id="speakable-summary" class="wp-block-paragraph">Enterprise AI company Cohere on Thursday launched its first voice model: Transcribe is an open source automatic speech recognition model that can be used for tasks like note-taking and speech analysis.</p>
<p class="wp-block-paragraph">Relatively light at just 2 billion parameters, the model is meant for use with consumer-grade GPUs for those who want to self-host it. It currently supports 14 languages: English, French, German, Italian, Spanish, Portuguese, Greek, Dutch, Polish, Chinese, Japanese, Korean, Vietnamese, and Arabic.</p>
<p class="wp-block-paragraph">Cohere says Transcribe beats models such as Zoom Scribe v1, IBM Granite 4.0 1B, ElevenLabs Scribe v2, and Qwen3-ASR-1.7B Speech on <a href="https://huggingface.co/spaces/hf-audio/open_asr_leaderboard" target="_blank" rel="noreferrer noopener nofollow">the Hugging Face Open ASR leaderboard</a>, achieving an average word error rate (WER) of 5.42, lower than any other model on the benchmark.</p>
<p class="wp-block-paragraph">The company claims Transcribe had an average win rate of 61% over other models when human evaluators assessed its transcriptions for accuracy, coherence, and usability. However, the model fell behind its rivals when it had to transcribe Portuguese, German, and Spanish.</p>
<p class="wp-block-paragraph">Cohere says Transcribe can process 525 minutes of audio in a minute, which is high for its class of model.</p>
<p class="wp-block-paragraph">The company is planning to integrate Transcribe into its enterprise agent orchestration platform, <a href="https://cohere.com/north" target="_blank" rel="noreferrer noopener nofollow">North</a>, and is making the model available through its <a href="http://dashboard.cohere.com/" target="_blank" rel="noreferrer noopener nofollow">API</a> for free. The model will also be available on <a href="https://cohere.com/solutions/model-vault" target="_blank" rel="noreferrer noopener nofollow">Model Vault</a>, Cohere’s managed inference platform.</p>
<p class="wp-block-paragraph">Speech recognition models are growing increasingly popular as demand grows for note-taking and dictation apps like Granola and <a href="https://techcrunch.com/2026/02/23/wispr-flow-launches-an-android-app-for-ai-powered-dictation/">Wispr Flow</a>.</p>
<p class="wp-block-paragraph">Earlier this year, Cohere reportedly <a href="https://www.cnbc.com/2026/02/13/ai-startup-cohere-revenue-ipo.html" target="_blank" rel="noreferrer noopener nofollow">told</a> investors that it was generating annual recurring revenue of $240 million in 2025, and its CEO, Aidan Gomez, was cited as saying that the startup <a href="https://techcrunch.com/2026/02/13/coheres-240m-year-sets-stage-for-ipo/">may go public “soon”</a>.</p>
© 2026 Winzheng.com 赢政天下 | 转载请注明来源并附原文链接