Decoder-only Architecture for Speech Recognition with CTC Prompts and Text Data Augmentation

TL;DR

This paper introduces a decoder-only architecture for speech recognition that leverages CTC prompts and text data augmentation, enabling effective use of text-only data and improving accuracy while maintaining computational efficiency.

Contribution

It proposes a novel decoder-only ASR model that utilizes CTC features as prompts and incorporates text augmentation, simplifying architecture and enhancing performance.

Findings

Reduced word error rates on LibriSpeech and Switchboard datasets.

Outperformed conventional encoder-decoder models in efficiency and accuracy.

Effective use of text-only data for speech recognition tasks.

Abstract

Collecting audio-text pairs is expensive; however, it is much easier to access text-only data. Unless using shallow fusion, end-to-end automatic speech recognition (ASR) models require architecture modifications or additional training schemes to use text-only data. Inspired by recent advances in decoder-only language models (LMs), such as GPT-3 and PaLM adopted for speech-processing tasks, we propose using a decoder-only architecture for ASR with simple text augmentation. To provide audio information, encoder features compressed by CTC prediction are used as prompts for the decoder, which can be regarded as refining CTC prediction using the decoder-only model. Because the decoder architecture is the same as an autoregressive LM, it is simple to enhance the model by leveraging external text data with LM training. An experimental comparison using LibriSpeech and Switchboard shows that our proposed models with text augmentation training reduced word error rates from ordinary CTC by 0.3% and 1.4% on LibriSpeech test-clean and testother set, respectively, and 2.9% and 5.0% on Switchboard and CallHome. The proposed model had advantage on computational efficiency compared with conventional encoder-decoder ASR models with a similar parameter setup, and outperformed them on the LibriSpeech 100h and Switchboard training scenarios.