Diarization-Guided Qwen-ASR Adaptation for Multilingual Two-Speaker Conversational Speech

This paper describes a system for the MLC-SLM 2026 Challenge that combines a modular speaker diarization front end with a challenge-adapted Qwen3-ASR-1.7B recognizer.

The diarization front end performs voice activity detection, subsegment generation, CAMPPlus speaker embedding extraction, two-speaker spectral clustering, and RTTM-based audio segmentation. The resulting speaker-attributed segments are grouped by language or region and decoded by the adapted ASR model. For ASR adaptation, the system first performs supervised full fine-tuning on official training data, then applies LoRA fine-tuning with synthetic speech from a three-pipeline TTS framework, and finally refines the model using GRPO reinforcement learning.

On the official development set, the full system achieves an average tcpMER of 23.70, reducing the error rate by 6.83 absolute points relative to the released Qwen-ASR-1.7B performance. On the final evaluation set, the system achieves an average tcpMER of 17.97.