Unsupervised Automatic Speech Recognition: A Review
arXiv:2106.04897 · doi:10.1016/j.specom.2022.02.005
Abstract
Automatic Speech Recognition (ASR) systems can be trained to achieve remarkable performance given large amounts of manually transcribed speech, but large labeled data sets can be difficult or expensive to acquire for all languages of interest. In this paper, we review the research literature to identify models and ideas that could lead to fully unsupervised ASR, including unsupervised segmentation of the speech signal, unsupervised mapping from speech segments to text, and semi-supervised models with nominal amounts of labeled examples. The objective of the study is to identify the limitations of what can be learned from speech data alone and to understand the minimum requirements for speech recognition. Identifying these limitations would help optimize the resources and efforts in ASR development for low-resource languages.
arXiv comment26 pages + 10 pages of references, 3 figures. Speech Communication (2022)
References in corpus (25)
- Efficient Estimation of Word Representations in Vector Space
- Representation Learning with Contrastive Predictive Coding
- Fast algorithm for detecting community structure in networks
- Deep Speech 2: End-to-End Speech Recognition in English and Mandarin
- A robust self-learning method for fully unsupervised cross-lingual mappings of word embeddings
- Libri-Light: A Benchmark for ASR with Limited or No Supervision
- Hierarchical Multiscale Recurrent Neural Networks
- Wav2Letter: an End-to-End ConvNet-based Speech Recognition System
- An Efficient, Probabilistically Sound Algorithm for Segmentation and Word Discovery
- End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architectures
- Representations of language in a model of visually grounded speech signal
- Effectiveness of self-supervised pre-training for speech recognition
- A segmental framework for fully-unsupervised large-vocabulary speech recognition
- End-to-End ASR-free Keyword Search from Speech
- Unsupervised Cross-Modal Alignment of Speech and Text Embedding Spaces
- An Unsupervised Autoregressive Model for Speech Representation Learning
- Unsupervised Speech Recognition
- The Zero Resource Speech Challenge 2020: Discovering discrete subword and word units
- Gate Activation Signal Analysis for Gated Recurrent Neural Networks and Its Correlation with Phoneme Boundaries
- A Fully Differentiable Beam Search Decoder
- Almost-unsupervised Speech Recognition with Close-to-zero Resource Based on Phonetic Structures Learned from Very Small Unpaired Speech and Text Data
- Unsupervised Sequence Classification using Sequential Output Statistics
- A Correspondence Variational Autoencoder for Unsupervised Acoustic Word Embeddings
- Analyzing the Robustness of Unsupervised Speech Recognition
- Unsupervised Speech Recognition via Segmental Empirical Output Distribution Matching