Deep LSTM Spoken Term Detection using Wav2Vec 2.0 Recognizer

TL;DR

This paper introduces a deep LSTM-based spoken term detection method utilizing Wav2Vec 2.0 recognizer, which effectively maps recognized speech and search terms into a shared embedding space, outperforming previous hybrid systems.

Contribution

It presents a novel approach combining Wav2Vec 2.0 with deep LSTM to improve spoken term detection without relying on pronunciation vocabularies.

Findings

Outperforms previous hybrid DNN-HMM systems on MALACH data

Effective transfer of pronunciation knowledge into grapheme-based Wav2Vec

Achieves significant accuracy improvements in English and Czech

Abstract

In recent years, the standard hybrid DNN-HMM speech recognizers are outperformed by the end-to-end speech recognition systems. One of the very promising approaches is the grapheme Wav2Vec 2.0 model, which uses the self-supervised pretraining approach combined with transfer learning of the fine-tuned speech recognizer. Since it lacks the pronunciation vocabulary and language model, the approach is suitable for tasks where obtaining such models is not easy or almost impossible. In this paper, we use the Wav2Vec speech recognizer in the task of spoken term detection over a large set of spoken documents. The method employs a deep LSTM network which maps the recognized hypothesis and the searched term into a shared pronunciation embedding space in which the term occurrences and the assigned scores are easily computed. The paper describes a bootstrapping approach that allows the transfer of the knowledge contained in traditional pronunciation vocabulary of DNN-HMM hybrid ASR into the context of grapheme-based Wav2Vec. The proposed method outperforms the previously published system based on the combination of the DNN-HMM hybrid ASR and phoneme recognizer by a large margin on the MALACH data in both English and Czech languages.