Lead2Gold: Towards exploiting the full potential of noisy transcriptions for speech recognition
The transcriptions used to train an Automatic Speech Recognition (ASR) system may contain errors. Usually, either a quality control stage discards transcriptions with too many errors, or the noisy transcriptions are used as is. We introduce Lead2Gold, a method to train an ASR system that exploits the full potential of noisy transcriptions. Based on a noise model of transcription errors, Lead2Gold searches for better transcriptions of the training data with a beam search that takes this noise model into account. The beam search is differentiable and does not require a forced alignment step, thus the whole system is trained end-to-end. Lead2Gold can be viewed as a new loss function that can be used on top of any sequence-to-sequence deep neural network. We conduct proof-of-concept experiments on noisy transcriptions generated from letter corruptions with different noise levels. We show that Lead2Gold obtains a better ASR accuracy than a competitive baseline which does not account for the (artificially-introduced) transcription noise.
Code (0)
등록된 구현이 없습니다.
Tasks
Automatic Speech RecognitionAutomatic Speech Recognition (ASR)speech-recognitionSpeech RecognitionSimilar Papers 제목 키워드 기반
An Analysis by Synthesis Method that Allows Accurate Spatial Modeling of Thickness of Cortical Bone from Clinical QCT
Osteoporosis is a skeletal disorder that leads to increased fracture risk due to decreased strength of cortical and trabecular bone. Even with state-of-the-art non-invasive assessment methods there is still a high underd…
Multi-Label Gold Asymmetric Loss Correction with Single-Label Regulators
Multi-label learning is an emerging extension of the multi-class classification where an image contains multiple labels. Not only acquiring a clean and fully labeled dataset in multi-label learning is extremely expensive…
Multi-class ClassificationMulti-Label LearningThe Curious Case of Factuality Finetuning: Models' Internal Beliefs Can Improve Factuality
Language models are prone to hallucination - generating text that is factually incorrect. Finetuning models on high-quality factual information can potentially reduce hallucination, but concerns remain; obtaining factual…
Enhanced Meta Label Correction for Coping with Label Corruption
Traditional methods for learning with the presence of noisy labels have successfully handled datasets with artificially injected noise but still fall short of adequately handling real-world noise. With the increasing use…
Image ClassificationLearning with noisy labelsTTM-RE: Memory-Augmented Document-Level Relation Extraction
Document-level relation extraction aims to categorize the association between any two entities within a document. We find that previous methods for document-level relation extraction are ineffective in exploiting the ful…
Document-level Relation ExtractionDocument-level RE with incomplete labelingRelationRelation Extraction