Search papers, labs, and topics across Lattice.
This paper presents a novel system for multilingual two-speaker conversational speech recognition, integrating a modular speaker diarization front end with the Qwen3-ASR-1.7B model. The approach employs a combination of supervised fine-tuning, LoRA fine-tuning with synthetic speech augmentation, and GRPO reinforcement learning to enhance model performance. The system achieves a significant reduction in error rates, with a tcpMER of 17.97 on the final evaluation set, demonstrating the effectiveness of the proposed methods in improving ASR robustness and accuracy.
Achieving a tcpMER of 17.97, this system reduces error rates significantly by leveraging advanced diarization and ASR adaptation techniques.
This paper describes our self-designed system for Task 1 of the MLC-SLM 2026 Challenge for multilingual two-speaker conversational speech. The system 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, we first perform supervised full fine-tuning on the official training data, then apply LoRA fine-tuning with synthetic speech generated by a three-pipeline TTS-based synthetic speech augmentation framework, and finally refine the model using GRPO reinforcement learning with rewards based on WER/CER and penalties for hallucination, repetition, and length deviation. 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. Ablation results show that supervised fine-tuning provides the largest gain, while synthetic-speech LoRA adaptation and reinforcement learning further improve robustness.
