智能眼镜语音识别系统在多人对话中表现优异，精准区分说话人并处理重叠语音。
The tttAI System for the TSA-ASR Task of the SmartGlasses Challenge 2026
- 分步式架构：先分说话人，再提取目标语音，最后识别。
- 两人对话准确率89.9%（CER 7.10%），多人会议34.04%。
- 适用于真实场景下的智能眼镜语音交互，适合语音处理研究者。
本文介绍了提交至SmartGlasses Challenge 2026 TSA-ASR任务的tttAI系统，该任务在两人对话（Track 1）和多人群会（Track 2）上进行评估，要求从智能眼镜录音中实现带时间戳的说话人标注语音识别。由于音频长、说话人多且频繁重叠，任务极具挑战性。系统采用级联架构，包括说话人分割、重叠检测、目标说话人提取、后处理和自动语音识别。分割模块使用WavLM-Large提取特征，通过Conformer编码器进行帧级分类，再以嵌入聚类生成全局说话人段。对重叠区域，采用WeSep-based目标说话人提取模型，结合ECAPA-TDNN说话人嵌入；当提取不可靠时，启用主说话人回退策略。最终系统使用FireRedASR2-AED模型，仅使用第一麦克风通道。系统总参数量约15.3亿。在Track 1上，tcpCER为7.10%；在Track 2上，tcpCER为34.04%，位列排行榜第二。
原文摘要 · Abstract (English)
This paper presents the tttAI system submitted to the TSA-ASR task of the SmartGlasses Challenge 2026, evaluated on both two-person dialogues (Track 1) and multi-party meetings (Track 2). The task requires time-stamped speaker-attributed speech recognition from smart-glasses recordings. This is particularly challenging due to long-form audio, multiple speakers, and frequent overlapping speech. We proposed a cascaded architecture consisting of speaker diarization, overlap detection, target-speaker extraction, post-processing, and automatic speech recognition. The diarization module extracts features via WavLM-Large, performs frame-wise speaker classification with a Conformer encoder, and then generates global speaker segments through embedding clustering. For overlapped regions, we apply a WeSep-based target-speaker extraction model with ECAPA-TDNN speaker embeddings. When the extraction is unreliable, a dominant-speaker fallback strategy is used. The final system uses FireRedASR2-AED with the first microphone channel. The submitted system has a total parameter count of approximately 1.53B. On Track 1, our system achieves a tcpCER of 7.10%. On Track 2, it achieves a tcpCER of 34.04% and ranks second on the leaderboard.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。