Negative transfer in training of acoustic models for automatic speech recognition has been reported in several contexts such as domain change or speaker characteristics. This paper …
In this work, we investigate a machine learning approach to cost-effectively train acoustic models for speech recognition. More specifically, we utilize an active learning method that …
T Tsutaoka, K Shinoda - Proceedings of The 2012 Asia Pacific …, 2012 - ieeexplore.ieee.org
We propose an acoustic model training method which combines committee-based active learning and semi-supervised learning for large vocabulary continuous speech recognition …
State-of-the-art speech recognition systems are trained using human transcriptions of speech utterances. In this paper, we describe a method to combine active and unsupervised …
F Lefevre, JL Gauvain, L Lamel - INTERSPEECH, 2001 - researchgate.net
Although there have been regular improvements in speech recognition technology over the past decade, speech recognition is far from being a solved problem. Recognition systems …
NF Chen, BP Lim… - Proceedings of the …, 2017 - experts.illinois.edu
It is challenging to obtain large amounts of native (matched) labels for audio in under- resourced languages. This could be due to a lack of literate speakers of the language or a …
F Lefevre, JL Gauvain, L Lamel - … ITRW) on Adaptation Methods for Speech …, 2001 - Citeseer
The last decade has witnessed major advances in core speech recognition technology, with today's systems able to recognize continuous speech from many speakers without the need …
Recently developed end-to-end (E2E) automatic speech recognition (ASR) systems demand abundance of transcribed speech data, there are several scenarios where the labeling of …
S Raghavan, K Shubham - Proc. Workshop on Machine …, 2021 - homepages.inf.ed.ac.uk
With the recent developments of highly parametric deep learning architectures, many recent works have demonstrated their effectiveness with End-to-End approaches in achieving state …