A Sidecar Separator Can Convert a Single-Talker Speech Recognition System to a Multi-Talker One
Abstract
Although automatic speech recognition (ASR) can perform well in common non-overlapping environments, sustaining performance in multi-talker overlapping speech recognition remains challenging. Recent research revealed that ASR model's encoder captures different levels of information with different layers -- the lower layers tend to have more acoustic information, and the upper layers more linguistic. This inspires us to develop a Sidecar separator to empower a well-trained ASR model for multi-talker scenarios by separating the mixed speech embedding between two suitable layers. We experimented with a wav2vec 2.0-based ASR model with a Sidecar mounted. By freezing the parameters of the original model and training only the Sidecar (8.7 M, 8.4% of all parameters), the proposed approach outperforms the previous state-of-the-art by a large margin for the 2-speaker mixed LibriMix dataset, reaching a word error rate (WER) of 10.36%; and obtains comparable results (7.56%) for LibriSpeechMix dataset when limited training.
Cite
@article{arxiv.2302.09908,
title = {A Sidecar Separator Can Convert a Single-Talker Speech Recognition System to a Multi-Talker One},
author = {Lingwei Meng and Jiawen Kang and Mingyu Cui and Yuejiao Wang and Xixin Wu and Helen Meng},
journal= {arXiv preprint arXiv:2302.09908},
year = {2023}
}
Comments
Accepted by IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2023