paper-with-me

Papers

Learning from Flawed Data: Weakly Supervised Automatic Speech Recognition

2023-09-26 · Dongji Gao, Hainan Xu, Desh Raj, Leibny Paola Garcia Perera, Daniel Povey, Sanjeev Khudanpur

Training automatic speech recognition (ASR) systems requires large amounts of well-curated paired data. However, human annotators usually perform "non-verbatim" transcription, which can result in poorly trained models. In this paper, we propose Omni-temporal Classification (OTC), a novel training criterion that explicitly incorporates label uncertainties originating from such weak supervision. This allows the model to effectively learn speech-text alignments while accommodating errors present in the training transcripts. OTC extends the conventional CTC objective for imperfect transcripts by leveraging weighted finite state transducers. Through experiments conducted on the LibriSpeech and LibriVox datasets, we demonstrate that training ASR models with OTC avoids performance degradation even with transcripts containing up to 70% errors, a scenario where CTC models fail completely. Our implementation is available at https://github.com/k2-fsa/icefall.

📄 PDF Abstract BibTeX arXiv:2309.15796

Code (1)

k2-fsa/icefall 공식 구현 pytorch

Tasks

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)speech-recognitionSpeech Recognition

Methods 이 논문이 사용한 방법론

fail 설명 없음

Similar Papers 제목 키워드 기반

Leveraging Weakly Supervised Data to Improve End-to-End Speech-to-Text Translation

2018-11-05 · Ye Jia, Melvin Johnson, Wolfgang Macherey, Ron J. Weiss 외

End-to-end Speech Translation (ST) models have many potential advantages when compared to the cascade of Automatic Speech Recognition (ASR) and text Machine Translation (MT) models, including lowered inference latency an…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Machine TranslationMulti-Task Learning+7

HebDB: a Weakly Supervised Dataset for Hebrew Speech Processing

2024-07-10 · Arnon Turetzky, Or Tal, Yael Segal-Feldman, Yehoshua Dissen 외

We present HebDB, a weakly supervised dataset for spoken language processing in the Hebrew language. HebDB offers roughly 2500 hours of natural and spontaneous speech recordings in the Hebrew language, consisting of a la…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)speech-recognitionSpeech Recognition

WST: Weakly Supervised Transducer for Automatic Speech Recognition

2025-11-06 · Dongji Gao, Chenda Liao, Changliang Liu, Matthew Wiesner 외 arxiv

The Recurrent Neural Network-Transducer (RNN-T) is widely adopted in end-to-end (E2E) automatic speech recognition (ASR) tasks but depends heavily on large-scale, high-quality annotated data, which are often costly and d…

Speech Recognition

Investigating self-supervised, weakly supervised and fully supervised training approaches for multi-domain automatic speech recognition: a study on Bangladeshi Bangla

2022-10-24 · Ahnaf Mozib Samin, M. Humayon Kobir, Md. Mushtaq Shahriyar Rafee, M. Firoz Ahmed 외

Despite huge improvements in automatic speech recognition (ASR) employing neural networks, ASR systems still suffer from a lack of robustness and generalizability issues due to domain shifting. This is mainly because pri…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)speech-recognitionSpeech Recognition+1

Weakly Supervised Construction of ASR Systems with Massive Video Data

2020-08-04 · Mengli Cheng, Chengyu Wang, Xu Hu, Jun Huang 외

Building Automatic Speech Recognition (ASR) systems from scratch is significantly challenging, mostly due to the time-consuming and financially-expensive process of annotating a large amount of audio data with transcript…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Optical Character RecognitionOptical Character Recognition (OCR)+3