Adaptation Algorithms for Neural Network-Based Speech Recognition: An Overview
| Autor: | Jinyu Li, Peter Bell, Steve Renals, Pawel Swietojanski, Ondrej Klejch, Joachim Fainberg |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
semi-supervised learning Domain adaptation Sound (cs.SD) Computer science domain adaptation Speech recognition 02 engineering and technology accent adaptation Computer Science - Sound 030507 speech-language pathology & audiology 03 medical and health sciences speaker embeddings Approximation error Audio and Speech Processing (eess.AS) Stress (linguistics) structured linear transforms 0202 electrical engineering electronic engineering information engineering FOS: Electrical engineering electronic engineering information engineering Adaptation (computer science) Hidden Markov model Computer Science - Computation and Language Artificial neural network speech recognition speaker adaptation TK1-9971 Accent adaptation regularization Model parameter Computer Science::Sound Signal Processing 020201 artificial intelligence & image processing Electrical engineering. Electronics. Nuclear engineering 0305 other medical science Focus (optics) Algorithm Computation and Language (cs.CL) Electrical Engineering and Systems Science - Audio and Speech Processing data augmentation |
| Zdroj: | IEEE Open Journal of Signal Processing, Vol 2, Pp 33-66 (2021) EEE Open Journal of Signal Processing Bell, P, Fainberg, J, Klejch, O, Li, J, Renals, S & Swietojanski, P 2021, ' Adaptation Algorithms for Neural Network-Based Speech Recognition: An Overview ', IEEE Open Journal of Signal Processing, vol. 2, pp. 33-66 . https://doi.org/10.1109/OJSP.2020.3045349 |
| ISSN: | 2644-1322 |
| Popis: | We present a structured overview of adaptation algorithms for neural network-based speech recognition, considering both hybrid hidden Markov model / neural network systems and end-to-end neural network systems, with a focus on speaker adaptation, domain adaptation, and accent adaptation. The overview characterizes adaptation algorithms as based on embeddings, model parameter adaptation, or data augmentation. We present a meta-analysis of the performance of speech recognition adaptation algorithms, based on relative error rate reductions as reported in the literature. Comment: Total of 31 pages, 27 figures. Associated repository: https://github.com/pswietojanski/ojsp_adaptation_review_2020 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|