Language-universal Speech Attributes Modeling For Zero-shot Multilingual Spoken Keyword Recognition

Abstract

We propose a novel language-universal approach to end-to-end automatic spoken keyword recognition (SKR) leveraging upon (i) a self-supervised pre-trained model, and (ii) a set of universal speech attributes (manner and place of articulation). Specifically, Wav2Vec2.0 is used to generate robust speech representations, followed by a linear output layer to produce attribute sequences. A non-trainable pronunciation model then maps sequences of attributes into spoken keywords in a multilingual setting. Experiments on the Multilingual Spoken Words Corpus show comparable performances to character- and phoneme-based SKR in seen languages. The inclusion of domain adversarial training (DAT) improves the proposed framework, outperforming both character- and phoneme-based SKR approaches with 13.73% and 17.22% relative word error rate (WER) reduction in seen languages, and achieves 32.14% and 19.92% WER reduction for unseen languages in zero-shot settings.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyyen2024language

Related papers

• Boosting End-to-end Multilingual Phoneme Recognition Through Exploiting Universal Speech Attributes Constraints (2023)5.84
• Phonmatchnet: Phoneme-guided Zero-shot Keyword Spotting For User-defined Keywords (2023)13.34
• Multilingual And Unsupervised Subword Modeling For Zero-resource Languages (2018)7.81
• Language-universal Adapter Learning With Knowledge Distillation For End-to-end Multilingual Speech Recognition (2023)0.95
• Unsupervised Speech Recognition (2021)0.00
• Massively Multilingual Adversarial Speech Recognition (2019)11.93
• Improved Language Identification Through Cross-lingual Self-supervised Learning (2021)10.61
• Improved Acoustic Word Embeddings For Zero-resource Languages Using Multilingual Transfer (2020)7.81