Analysis of low-resource acoustic model self-training

By Scott Novotney, Richard M. Schwartz. 2009

26 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Speech & Audio, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Yu Zhang
2. Jinyu Li
3. Gabriel Synnaeve
4. Yonghui Wu
5. Quoc V. Le
6. Wei-Ning Hsu
7. Michael Auli
8. Ronan Collobert

In Speech & Audio

1. Yu Zhang
2. Jinyu Li
3. Wei-Ning Hsu
4. Gabriel Synnaeve
5. Yonghui Wu
6. Chung-Cheng Chiu
7. Alexei Baevski
8. Ronan Collobert

In Signal Processing

1. Yu Zhang
2. Jinyu Li
3. Yonghui Wu
4. Chung-Cheng Chiu
5. Gabriel Synnaeve
6. Ye Jia
7. Ronan Collobert
8. Ruoming Pang

Citations in the corpus

• Pushing the Limits of Semi-Supervised Learning for Automatic Speech Recognition (2020, methodology): “Self-training has been a popular technique in ASR studied extensively in the literature [25–33].”
• Improved Noisy Student Training for Automatic Speech Recognition (2020, background): “Self-training has a long history of being applied to ASR through numerous studies, e.g., [12, 13, 14, 15, 16, 17, 18, 19].”
• w2v-BERT: Combining Contrastive Learning and Masked Language Modeling for Self-Supervised Speech Pre-Training (2021, methodology): “Self-training has been a practically useful and extensively studied technique in ASR [4, 5, 6, 7, 8, 9].”
• slimIPL: Language-Model-Free Iterative Pseudo-Labeling (2020, background): “o-labeled datasets. Many recent works on self-training propose and validate these extensions, including those in computer vision [9, 10], natural language processing [11, 12, 13, 14, 15, 16, 17], ASR [18, 19, 20, 7, 8], and speech translation [21]. An extension to the simple pseudo-labeling method c”
• Semi-Supervised DNN Training with Word Selection for ASR (2017, methodology): “Whereas in [4, 5] the GMM-HMM training data are selected per-sentence, choosing the sentences which are believed to have the WER smaller than the overall WER of the development set.”
• Deep neural network features and semi-supervised training for low resource speech recognition (2013, unknown): “Semi-supervised training has been effectively used to train acoustic models in several languages and conditions [32, 33, 34, 35, 36].”
• Semi-supervised training of Deep Neural Networks (2013, methodology): “This corresponds to WER recovery [2] of 36%.”