Introduction to statistical pattern recognition (2nd ed.)
Introduction to statistical pattern recognition (2nd ed.)
Language-independent and language-adaptive acoustic modeling for speech recognition
Speech Communication
Development of Dialect-Specific Speech Recognizers Using Adaptation Methods
ICASSP '97 Proceedings of the 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP '97)-Volume 2 - Volume 2
Tree-based state tying for high accuracy acoustic modelling
HLT '94 Proceedings of the workshop on Human Language Technology
Language-dependent state clustering for multilingual acoustic modelling
Speech Communication
Multidialectal Spanish acoustic modeling for speech recognition
Speech Communication
| Hi-index | 0.00 |
Although English is spoken throughout South Africa it is most often used as a second or third language, resulting in several prevalent accents within the same population. When dealing with multiple accents in this under-resourced environment, automatic speech recognition (ASR) is complicated by the need to compile multiple, accent-specific speech corpora. We investigate how best to combine speech data from five South African accents of English in order to improve overall speech recognition performance. Three acoustic modelling approaches are considered: separate accent-specific models, accent-independent models obtained by pooling training data across accents, and multi-accent models. The latter approach extends the decision-tree clustering process normally used to construct tied-state hidden Markov models (HMMs) by allowing questions relating to accent. We find that multi-accent modelling outperforms accent-specific and accent-independent modelling in both phone and word recognition experiments, and that these improvements are statistically significant. Furthermore, we find that the relative merits of the accent-independent and accent-specific approaches depend on the particular accents involved. Multi-accent modelling therefore offers a mechanism by which speech recognition performance can be optimised automatically, and for hard decisions regarding which data to pool and which to separate to be avoided.