Large-Scale Semi-Supervised Training in Deep Learning Acoustic Model for ASR

Long, Yanhua; Li, Yijie; Wei, Shuang; Zhang, Qiaozheng; Yang, Chang‐Fa

doi:10.1109/access.2019.2940961

Cited by 20 publications

(7 citation statements)

References 37 publications

(67 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Researchers have explored different methods of semi-supervised training for speech recognition. Long et al (2019) investigate large-scale semisupervised training to improve acoustic models for automatic speech recognition. They provide an empirical analysis of semi-supervised training with respect to transcription quality, data quality, filtering, etc.…”

Section: Related Workmentioning

confidence: 99%

Improving End-to-End Bangla Speech Recognition with Semi-supervised Training

Sadeq

Chowdhury²,

Utshaw³

et al. 2020

Findings of the Association for Computational Linguistics: EMNLP 2020

View full text Add to dashboard Cite

Automatic speech recognition systems usually require large annotated speech corpus for training. The manual annotation of a large corpus is very difficult. It can be very helpful to use unsupervised and semi-supervised learning methods in addition to supervised learning. In this work, we focus on using a semi-supervised training approach for Bangla Speech Recognition that can exploit large unpaired audio and text data. We encode speech and text data in an intermediate domain and propose a novel loss function based on the global encoding distance between encoded data to guide the semisupervised training. Our proposed method reduces the Word Error Rate (WER) of the system from 37% to 31.9%.

show abstract

Section: Related Workmentioning

confidence: 99%

Improving End-to-End Bangla Speech Recognition with Semi-supervised Training

Sadeq

Chowdhury²,

Utshaw³

et al. 2020

Findings of the Association for Computational Linguistics: EMNLP 2020

View full text Add to dashboard Cite

show abstract

“…One stream of works applies unsupervised/semi-supervised methods to tackle the problem. For example, Long et al [21] propose an improved self-training approach for semi-supervised training of DNN and RNN based acoustic models. Karita et al [22] jointly learn ASR and TTS models for semi-supervised training of ASR models.…”

Section: Related Workmentioning

confidence: 99%

Weakly Supervised Construction of ASR Systems with Massive Video Data

Cheng¹,

Wang²,

Huang³

et al. 2020

Preprint

View full text Add to dashboard Cite

Building Automatic Speech Recognition (ASR) systems from scratch is significantly challenging, mostly due to the timeconsuming and financially-expensive process of annotating a large amount of audio data with transcripts. Although several unsupervised pre-training models have been proposed, applying such models directly might still be sub-optimal if more labeled, training data could be obtained without a large cost.In this paper, we present a weakly supervised framework for constructing ASR systems with massive video data. As videos often contain human-speech audios aligned with subtitles, we consider videos as an important knowledge source, and propose an effective approach to extract high-quality audios aligned with transcripts from videos based on Optical Character Recognition (OCR). The underlying ASR model can be fine-tuned to fit any domain-specific target training datasets after weakly supervised pre-training. Extensive experiments show that our framework can easily produce state-of-the-art results on six public datasets for Mandarin speech recognition. 1

show abstract

“…In this work, we focus on this setting to explore how much gain can be made without undertaking the expensive and tedious task of annotating accented speech. Several semi-supervised learning techniques have been proposed to tackle this scenario by combining annotated and unannotated examples to jointly learn robust inference models [5][6][7]. We propose to leverage domain adversarial training * Work conducted as an intern at Amazon AWS AI.…”

Section: Introductionmentioning

confidence: 99%

Best of Both Worlds: Robust Accented Speech Recognition with Adversarial Transfer Learning

Das¹,

Bodapati²,

Sunkara³

et al. 2021

Preprint

View full text Add to dashboard Cite

Training deep neural networks for automatic speech recognition (ASR) requires large amounts of transcribed speech. This becomes a bottleneck for training robust models for accented speech which typically contains high variability in pronunciation and other semantics, since obtaining large amounts of annotated accented data is both tedious and costly. Often, we only have access to large amounts of unannotated speech from different accents. In this work, we leverage this unannotated data to provide semantic regularization to an ASR model that has been trained only on one accent, to improve its performance for multiple accents. We propose Accent Pre-Training (Acc-PT), a semi-supervised training strategy that combines transfer learning and adversarial training. Our approach improves the performance of a state-of-the-art ASR model by 33% on average over the baseline across multiple accents, training only on annotated samples from one standard accent, and as little as 105 minutes of unannotated speech from a target accent.

show abstract

Large-Scale Semi-Supervised Training in Deep Learning Acoustic Model for ASR

Cited by 20 publications

References 37 publications

Improving End-to-End Bangla Speech Recognition with Semi-supervised Training

Improving End-to-End Bangla Speech Recognition with Semi-supervised Training

Weakly Supervised Construction of ASR Systems with Massive Video Data

Best of Both Worlds: Robust Accented Speech Recognition with Adversarial Transfer Learning

Contact Info

Product

Resources

About