发表机构
Ant Group; UESTC（蚂蚁集团; 电子科技大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对第二届MLC-SLM挑战赛的两项任务，分别采用前置静音增强等策略微调VibeVoice-ASR-7B、用合成问答对微调Qwen3-Omni-30B-A3B-Instruct，提升了任务1和任务2的评估性能。
AI 中文摘要
第二届多语种会话语音语言模型（MLC-SLM）挑战赛针对完整、未分段的多语种会话评估两项任务：说话人 diarization（说话人分割）与识别（任务1）、会话语音理解（任务2）。两项任务在评估时均不提供 oracle 话语边界或说话人标签，且任务2无问答训练集。针对任务1，我们通过随机前置静音裁剪、一致时间戳修正及指数移动平均（EMA）训练策略微调VibeVoice-ASR-7B；针对任务2，我们通过多模态候选生成、静音音频过滤及分布匹配增强构建合成问答对，并微调Qwen3-Omni-30B-A3B-Instruct以实现带标签的直接回答。在任务1评估集上，裁剪将tcpMER从18.30%降至17.27%，EMA进一步将其降至16.73%；在任务2评估集上，联合应用分布匹配增强与带标签直接回答将准确率从83.0%提升至86.0%。
英文摘要
The second Multilingual Conversational Speech Language Model (MLC-SLM) Challenge evaluates two tasks over complete, unsegmented multilingual conversations: speaker diarization and recognition (Task 1) and conversational speech understanding (Task 2). Neither task provides oracle utterance boundaries or speaker labels at evaluation, and Task 2 provides no question-answer training set. For Task 1, we fine-tune VibeVoice-ASR-7B with random leading-silence cropping, consistent timestamp correction, and an exponential moving average (EMA) training strategy. For Task 2, we construct synthetic question-answer pairs through multimodal candidate generation, silent-audio filtering, and distribution-matched augmentation, and fine-tune Qwen3-Omni-30B-A3B-Instruct for tagged direct answering. On the Task 1 evaluation set, cropping reduces tcpMER from 18.30% to 17.27%, and EMA further reduces it to 16.73%. On the Task 2 evaluation set, jointly applying distribution-matched augmentation and tagged direct answering raises accuracy from 83.0% to 86.0%.
