2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information

2021 IEEE International Conference on Acoustics, Speech and Signal Processing

6-11 June 2021 • Toronto, Ontario, Canada

Extracting Knowledge from Information
Login Paper Search My Schedule Paper Index Help

My ICASSP 2021 Schedule

Note: Your custom schedule will not be saved unless you create a new account or login to an existing account.
  1. Create a login based on your email (takes less than one minute)
  2. Perform 'Paper Search'
  3. Select papers that you desire to save in your personalized schedule
  4. Click on 'My Schedule' to see the current list of selected papers
  5. Click on 'Printable Version' to create a separate window suitable for printing (the header and menu will appear, but will not actually print)

Paper Detail

Paper IDHLT-9.4
Paper Title GENERATING HUMAN READABLE TRANSCRIPT FOR AUTOMATIC SPEECH RECOGNITION WITH PRE-TRAINED LANGUAGE MODEL
Authors Junwei Liao, University of Electronic Science and Technology of China, China; Yu Shi, Microsoft Cognitive Services Research Group, United States; Ming Gong, Linjun Shou, Microsoft STCA NLP Group, China; Sefik Eskimez, Liyang Lu, Microsoft Cognitive Services Research Group, United States; Hong Qu, University of Electronic Science and Technology of China, China; Michael Zeng, Microsoft Cognitive Services Research Group, United States
SessionHLT-9: Style and Text Normalization
LocationGather.Town
Session Time:Wednesday, 09 June, 16:30 - 17:15
Presentation Time:Wednesday, 09 June, 16:30 - 17:15
Presentation Poster
Topic Speech Processing: [SPE-GASR] General Topics in Speech Recognition
IEEE Xplore Open Preview  Click here to view in IEEE Xplore
Abstract Modern Automatic Speech Recognition (ASR) systems can achieve high performance in terms of recognition accuracy. However, a perfectly accurate transcript still can be challenging to read due to disfluency, filter words, and other errata common in spoken communication. Many downstream tasks and human readers rely on the output of the ASR system; therefore, errors introduced by the speaker and ASR system alike will be propagated to the next task in the pipeline. In this work, we propose an ASR post-processing model that aims to transform the incorrect and noisy ASR output into a readable text for humans and downstream tasks. We leverage the Metadata Extraction (MDE) corpus to construct a task-specific dataset for our study. Since the dataset is small, we propose a novel data augmentation method and use a two-stage training strategy to fine-tune the RoBERTa pre-trained model. On the constructed test set, our model outperforms a production two-step pipeline-based post-processing method by a large margin of 13.26 on readability-aware WER (RA-WER) and 17.53 on BLEU metrics. Human evaluation also demonstrates that our method can generate more human-readable transcripts than the baseline method.