{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/iterative-pseudo-labeling-for-speech","title":"Iterative Pseudo-Labeling for Speech Recognition","arxiv_id":"2005.09267","date":"2020-05-19","proceeding":null,"authors":["Qiantong Xu","Tatiana Likhomanenko","Jacob Kahn","Awni Hannun","Gabriel Synnaeve","Ronan Collobert"],"abstract":"Pseudo-labeling has recently shown promise in end-to-end automatic speech recognition (ASR). We study Iterative Pseudo-Labeling (IPL), a semi-supervised algorithm which efficiently performs multiple iterations of pseudo-labeling on unlabeled data as the acoustic model evolves. In particular, IPL fine-tunes an existing model at each iteration using both labeled data and a subset of unlabeled data. We study the main components of IPL: decoding with a language model and data augmentation. We then demonstrate the effectiveness of IPL by achieving state-of-the-art word-error rate on the Librispeech test sets in both standard and low-resource setting. We also study the effect of language models trained on different corpora to show IPL can effectively utilize additional text. Finally, we release a new large in-domain text corpus which does not overlap with the Librispeech training transcriptions to foster research in low-resource, semi-supervised ASR","url_abs":"https://arxiv.org/abs/2005.09267v2","url_pdf":"https://arxiv.org/pdf/2005.09267v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"iterative-pseudo-labeling-for-speech","repo_url":"https://github.com/facebookresearch/wav2letter/tree/master/recipes/ipl","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"automatic-speech-recognition-2","task_name":"Automatic Speech Recognition"},{"task_slug":"automatic-speech-recognition","task_name":"Automatic Speech Recognition (ASR)"},{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"}],"methods":[{"method_slug":"ipl","method_name":"IPL"}],"datasets_introduced":[],"methods_introduced":[{"slug":"ipl","name":"IPL","full_name":"Iterative Pseudo-Labeling"}],"results":[{"leaderboard":"/sota/speech-recognition-on-librispeech-test-clean","task":"Speech Recognition","dataset":"LibriSpeech test-clean","model":"Conv + Transformer AM + Iterative Pseudo-Labeling (n-gram LM + Transformer Rescoring)","rank_in_archive_order":28,"of":64,"metrics":{"Word Error Rate (WER)":"2.10"},"uses_additional_data":false},{"leaderboard":"/sota/speech-recognition-on-librispeech-test-other","task":"Speech Recognition","dataset":"LibriSpeech test-other","model":"Conv + Transformer AM + Iterative Pseudo-Labeling (n-gram LM + Transformer Rescoring)","rank_in_archive_order":13,"of":53,"metrics":{"Word Error Rate (WER)":"3.83"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2005.09267","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}