Who Needs Words? Lexicon-Free Speech Recognition

Abstract

Lexicon-free speech recognition naturally deals with the problem of out-of-vocabulary (OOV) words. In this paper, we show that character-based language models (LM) can perform as well as word-based LMs for speech recognition, in word error rates (WER), even without restricting the decoding to a lexicon. We study character-based LMs and show that convolutional LMs can effectively leverage large (character) contexts, which is key for good speech recognition performance downstream. We specifically show that the lexicon-free decoding performance (WER) on utterances with OOV words using character-based LMs is better than lexicon-based decoding, both with character or word-based LMs.

8 pages, 1 figure

References in corpus (9)

• On the difficulty of training Recurrent Neural Networks
• Convolutional Sequence to Sequence Learning
• Improved training of end-to-end attention models for speech recognition
• Wav2Letter: an End-to-End ConvNet-based Speech Recognition System
• wav2letter++: The Fastest Open-source Speech Recognition System
• Fully Convolutional Speech Recognition
• The CAPIO 2017 Conversational Speech Recognition System
• Towards End-to-End Speech Recognition with Deep Convolutional Neural Networks
• Deep Recurrent Neural Networks for Acoustic Modelling

Cited by in corpus (9)

• vq-wav2vec: Self-Supervised Learning of Discrete Speech Representations
• End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architectures
• Espresso: A Fast End-to-end Neural Speech Recognition Toolkit
• Emphasizing Unseen Words: New Vocabulary Acquisition for End-to-End Speech Recognition
• HarperValleyBank: A Domain-Specific Spoken Dialog Corpus
• General-Purpose Speech Representation Learning through a Self-Supervised Multi-Granularity Framework
• End-to-end Whispered Speech Recognition with Frequency-weighted Approaches and Pseudo Whisper Pre-training
• PyChain: A Fully Parallelized PyTorch Implementation of LF-MMI for End-to-End ASR
• G2G: TTS-Driven Pronunciation Learning for Graphemic Hybrid ASR