NVIDIA 新研究：语音模型工具调用性能达 92% 准确率
NVIDIA 新研究，语音模型工具调用准确率 92%，比 GPT-5 在文本模式下的 85% 还高一点，解决语音代理性能瓶颈。
NVIDIA 的最新研究展示了全双工语音模型在客服任务中的表现。在干净环境下，商业双工语音模型完成 31% 到 51% 的任务，而 GPT-5 在文本模式下能达到 85%。研究指出语音代理性能下降主要源于语音处理流程。解决方案是将决策从语音模型中移出，让前端学习发出委托令牌，将转录文本转发给文本后端 LLM 进行工具调用，再通过轻量级预填充和重复机制接收结果并流式合成语音输出。
NVIDIA research papers are on fire recently! Here is another interesting paper where they give full-duplex speech models tool calls. (bookmark it) Commercial duplex voice models complete 31 to 51 percent of grounded customer-service tasks under clean conditions. Text agents like GPT-5 reach 85 percent on the same tasks in text mode. Most of what a voice agent loses, it loses in the speech pipeline. The fix routes the decision out of the speech model. The duplex frontend learns to emit a delegation token, forwards streaming transcripts to a text backend LLM for the tool call, and receives the result through a lightweight prefill-and-repeat mechanism before streaming TTS speaks it. Tool-call recall runs 92.0 to 97.2 percent with 81.2 percent accuracy at rejecting irrelevant calls. Turn-taking rate, streaming ASR word error rate and spoken-language intelligence all stay at the no-tool-call baseline. Paper: arxiv.org/abs/2609.19334 Chat with Paper: academy.dair.ai/papers/a-front… 💬 6 🔄 1 ❤️ 9 👀 1880 📊 9 ⚡
