中文标题#
傳音多語言語音識別系統用於 MLC-SLM 2025 挑戰賽
英文标题#
Transsion Multilingual Speech Recognition System for MLC-SLM 2025 Challenge
中文摘要#
本文介紹了由傳音語音團隊為 MLC-SLM 2025 挑戰賽的 Track 1 開發的新型多語言自動語音識別（ASR）系統的架構和性能。該系統包含三個關鍵組件：1）基於凍結的 Whisper-large-v3 的語音編碼器，利用大規模預訓練確保穩健的聲學特徵提取；2）使用 Linear-ReLU-Linear 變換機制的可訓練適配模塊，以有效對齊語音和文本表示；以及 3）與可訓練 LoRA 集成的凍結 Qwen2.5-7B-Instruct 大語言模型（LLM），用於優化上下文語言解碼。通過系統地結合預訓練模型與任務特定微調，該系統在評估集的 11 種語言中實現了 9.83% 的詞 / 字符錯誤率（WER/CER），並在全球參與者中排名第三。
英文摘要#
This paper presents the architecture and performance of a novel Multilingual Automatic Speech Recognition (ASR) system developed by the Transsion Speech Team for Track 1 of the MLC-SLM 2025 Challenge. The proposed system comprises three key components: 1) a frozen Whisper-large-v3 based speech encoder, leveraging large-scale pretraining to ensure robust acoustic feature extraction; 2) a trainable adaptor module using Linear-ReLU-Linear transformation mechanisms to effectively align speech and text representations; and 3) a frozen Qwen2.5-7B-Instruct large language model (LLM) integrated with trainable LoRA for optimized contextual linguistic decoding. By systematically combining pretrained models with task specific fine-tuning, the system achieved a word/character error rate (WER/CER) of 9.83% across 11 languages in the evaluation set and ranked third place among global participants.
文章页面#
傳音多語言語音識別系統用於 MLC-SLM 2025 挑戰賽
PDF 获取#
抖音掃碼查看更多精彩內容