Unsupervised training and directed manual transcription for LVCSR

By Kai Yu, Mark John Francis Gales, Lan Wang, Philip C. Woodland. 2010

80 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. Jinyu Li
2. Dong Yu
3. Tara N. Sainath
4. Li Deng
5. Shinji Watanabe
6. Daniel Povey
7. Sanjeev Khudanpur
8. Yu Zhang

In Speech & Audio

1. Jinyu Li
2. Tara N. Sainath
3. Sanjeev Khudanpur
4. Daniel Povey
5. Dong Yu
6. Shinji Watanabe
7. Vimal Manohar
8. Bo Li

In Signal Processing

1. Jinyu Li
2. Dong Yu
3. Shinji Watanabe
4. Daniel Povey
5. Sanjeev Khudanpur
6. Tara N. Sainath
7. Li Deng
8. Yifan Gong

Citations in the corpus

• Iterative Pseudo-Labeling for Speech Recognition (2020, neutral): “Semi-supervision in ASR is well-studied [15, 16, 17, 18, 19, 20]; our work builds primarily on recent work with end-to-end systems, especially PL [9, 1].”
• Semi-Supervised Training of Acoustic Models Using Lattice-Free MMI (2018, background): “Discriminative training is sensitive to the accuracy of the transcripts [10, 11, 12].”
• Semi-supervised maximum mutual information training of deep neural network acoustic models (2015, background): “Discriminative training is very sensitive to the accuracy of the transcripts [17, 18, 19].”
• Semi-supervised ensemble DNN acoustic model training (2017, methodology): “The most widely-used method is the utterance-level selection, which sorts the utterances by an utterance-level confidence measure score (CMS) and selects a certain percentage of top utterances for model training [5].”
• Semi-supervised Training for End-to-end Models via Weak Distillation (2019, methodology): “Confidence scores from an existing ASR system is commonly used to select unsupervised data for training with more data [11–15].”
• A big data approach to acoustic model training corpus selection (2014, methodology): “Rather than relying on small supervised training sets, one can instead exploit the large amount of audio data that is processed by systems like Voice Search, together with semi-supervised techniques to construct large training sets with automatically derived transcripts [4, 5, 6, 7, 8].”
• Semi-supervised GMM and DNN acoustic model training with multi-system combination and confidence re-calibration (2013, background): “For this reason, much effort has been devoted to the unsupervised and semi-supervised acoustic model training [1, 2, 3, 4, 5, 6, 7, 8] to exploit the untranscribed data.”