Denoising GER: A Noise-Robust Generative Error Correction with LLM for Speech Recognition
By: Yanyan Liu , Minqiang Xu , Yihao Chen and more
Potential Business Impact:
Fixes speech errors in noisy places.
In recent years, large language models (LLM) have made significant progress in the task of generation error correction (GER) for automatic speech recognition (ASR) post-processing. However, in complex noisy environments, they still face challenges such as poor adaptability and low information utilization, resulting in limited effectiveness of GER. To address these issues, this paper proposes a noise-robust multi-modal GER framework (Denoising GER). The framework enhances the model's adaptability to different noisy scenarios through a noise-adaptive acoustic encoder and optimizes the integration of multi-modal information via a heterogeneous feature compensation dynamic fusion (HFCDF) mechanism, improving the LLM's utilization of multi-modal information. Additionally, reinforcement learning (RL) training strategies are introduced to enhance the model's predictive capabilities. Experimental results demonstrate that Denoising GER significantly improves accuracy and robustness in noisy environments and exhibits good generalization abilities in unseen noise scenarios.
Similar Papers
Listening and Seeing Again: Generative Error Correction for Audio-Visual Speech Recognition
Multimedia
Helps computers understand speech by seeing and hearing.
Incorporating Error Level Noise Embedding for Improving LLM-Assisted Robustness in Persian Speech Recognition
Computation and Language
Helps computers understand noisy speech better.
DLRREC: Denoising Latent Representations via Multi-Modal Knowledge Fusion in Deep Recommender Systems
Information Retrieval
Makes movie suggestions much smarter.