Improving Semi-supervised End-to-end Automatic Speech Recognition using CycleGAN and Inter-domain Losses
Not on the map. Why?
Abstract (arXiv)
We propose a novel method that combines CycleGAN and inter-domain losses for semi-supervised end-to-end automatic speech recognition. Inter-domain loss targets the extraction of an intermediate shared representation of speech and text inputs using a shared network. CycleGAN uses cycle-consistent loss and the identity mapping loss to preserve relevant characteristics of the input feature after converting from one domain to another. As such, both approaches are suitable to train end-to-end models on unpaired speech-text inputs. In this paper, we exploit the advantages from both inter-domain loss and CycleGAN to achieve better shared representation of unpaired speech and text inputs and thus improve the speech-to-text mapping. Our experimental results on the WSJ eval92 and Voxforge (non English) show 8~8.5% character error rate reduction over the baseline, and the results on LibriSpeech test_clean also show noticeable improvement.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 22 of its 35 references that are in the map. First posted on arXiv 20 Oct 2022.
Builds on
22 of 35 references.
References · 22
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
- 2022Generative Adversarial Networkscited by 5,603
- 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,530
- 2017Unsupervised Image-to-Image Translation Networkscited by 602
- 2018ESPnet: End-to-End Speech Processing Toolkitcited by 554
- 2015Listen, Attend and Spellcited by 542
- 2015Deep Speech 2: End-to-End Speech Recognition in English and Mandarincited by 449
- 2015Attention-Based Models for Speech Recognitioncited by 400
- 2014Deep Speech: Scaling up end-to-end speech recognitioncited by 354
- 2017Unsupervised Machine Translation Using Monolingual Corpora Onlycited by 272
- 2016Joint CTC-Attention based End-to-End Speech Recognition using Multi-task Learningcited by 254
- 2015End-to-End Attention-based Large Vocabulary Speech Recognitioncited by 148
10 more
- 2017Unsupervised Neural Machine Translationcited by 148
- 2015EESEN: End-to-End Speech Recognition using Deep RNN Models and WFST-based Decodingcited by 73
- 2014End-to-end Continuous Speech Recognition using Attention-based Recurrent NN: First Resultscited by 68
- 2017Listening while Speaking: Speech Chain by Deep Learningcited by 46
- 2016Very Deep Convolutional Networks for End-to-End Speech Recognitioncited by 34
- 2016Neural Speech Recognizer: Acoustic-to-Word LSTM Model for Large Vocabulary Speech Recognitioncited by 33
- 2018Cycle-consistency training for end-to-end speech recognitioncited by 31
- 2018Back-Translation-Style Data Augmentation for End-to-End ASRcited by 28
- 2019Semi-supervised Sequence-to-sequence ASR using Unpaired Speech and Textcited by 10
- 2018Pretraining by Backtranslation for End-to-end ASR in Low-Resource Settingscited by 7
Built on by
Citing papers per year: 2024: 1.
1 paper.