Leveraging Language Information for Target Language Extraction

TL;DR

This paper introduces a new end-to-end framework that leverages speech pre-trained models to improve target language extraction from multilingual audio mixtures, demonstrating significant performance gains.

Contribution

It proposes a novel approach that uses language knowledge from pre-trained models to enhance extraction accuracy, and provides the first multilingual dataset for this task.

Findings

Achieves over 1.2 dB SI-SNR improvement for English and German extraction.

Constructs the first publicly available multilingual dataset for Target Language Extraction.

Demonstrates the effectiveness of language knowledge guidance in speech extraction.

Abstract

Target Language Extraction aims to extract speech in a specific language from a mixture waveform that contains multiple speakers speaking different languages. The human auditory system is adept at performing this task with the knowledge of the particular language. However, the performance of the conventional extraction systems is limited by the lack of this prior knowledge. Speech pre-trained models, which capture rich linguistic and phonetic representations from large-scale in-the-wild corpora, can provide this missing language knowledge to these systems. In this work, we propose a novel end-to-end framework to leverage language knowledge from speech pre-trained models. This knowledge is used to guide the extraction model to better capture the target language characteristics, thereby improving extraction quality. To demonstrate the effectiveness of our proposed approach, we construct the first publicly available multilingual dataset for Target Language Extraction. Experimental results show that our method achieves improvements of 1.22 dB and 1.12 dB in SI-SNR for English and German extraction, respectively, from mixtures containing both languages.