📄 中文摘要
Cohere发布了Transcribe，这是一款新的开源自动语音识别（ASR）模型，拥有20亿个参数。该模型以5.42%的平均词错误率（WER）在Hugging Face开源ASR排行榜上名列第一，超越了OpenAI的Whisper Large v3、ElevenLabs的Scribe v2以及阿里巴巴的Qwen3-ASR-1.7B等竞争对手。Cohere表示，Transcribe在语音识别领域的表现将为开发者和研究人员提供更强大的工具。
Cohere Transcribe：2B参数开源ASR模型实现5.42% WER，登顶Hugging Face排行榜
📄 English Summary
Cohere Transcribe: 2B-Parameter Open-Source ASR Model Achieves 5.42% WER, Topping Hugging Face Leaderboard
Cohere has released Transcribe, a new open-source automatic speech recognition (ASR) model featuring 2 billion parameters. It claims the top position on the Hugging Face Open ASR Leaderboard with an average word error rate (WER) of 5.42%, outperforming established competitors such as OpenAI's Whisper Large v3, ElevenLabs Scribe v2, and Alibaba's Qwen3-ASR-1.7B. According to Cohere, Transcribe's performance in speech recognition will provide developers and researchers with more powerful tools.