NVIDIA 发布 Nemotron 3.5 Transcribe 开源语音转写模型
NVIDIA 的开源语音转写模型用 4B 激活参数做到 2.8% 错误率，比 Mistral Voxtral Small 少 6 倍参数还打平，做语音智能体的可以看看。
NVIDIA 推出开源语音转写模型 Nemotron 3.5 Transcribe，总参数量 30B、激活参数仅 4B。该模型在 AA-WER Index 上取得 2.8% 的错误率，与 Mistral 的 Voxtral Small 并列开源模型第一。在语音智能体对话数据集 AA-AgentTalk 上，它以 1.8% 的错误率领先 Voxtral Small 的 2.3%。相比 Voxtral Small 的 24B 激活参数，它用少 6 倍的激活参数达到同等准确率，而同规模开源模型的错误率普遍在 4% 左右。
NVIDIA's Nemotron 3.5 Transcribe ties for #1 among open weights Speech to Text models with 2.8% WER, using just 4B active parameters
Nemotron 3.5 Transcribe is NVIDIA's new open weights Speech to Text model for non-streaming transcription, with 30B total and 4B active parameters, and the most accurate NVIDIA model on our non-streaming Speech to Text leaderboard.
Key takeaways
➤ Accuracy: Nemotron 3.5 Transcribe scores 2.8% on the AA-WER Index, tied with Mistral's Voxtral Small for #1 among open-weight models. It leads Voxtral Small on AA-AgentTalk, our dataset of voice agent conversations, at 1.8% vs. 2.3%.
➤ Model size: With 4B active parameters (30B total), Nemotron 3.5 Transcribe sits on the open weights frontier of AA-WER vs. active parameters. It matches the accuracy of Mistral's Voxtral Small with 6x fewer active parameters (4B vs. 24B), while other open weights models of a similar size sit at around 4% WER.
Congratulations to @NVIDIAAI on the release!
See more details below ⬇️
