Towards An ASR Approach Using Acoustic and Language Models for Speech Enhancement.
Khandokar Md. NayemDonald S. WilliamsonPublished in: ICASSP (2021)
Keyphrases
- language model
- speech recognition
- noisy speech
- speech enhancement
- automatic speech recognition
- speech signal
- noisy environments
- acoustic features
- language modeling
- n gram
- word error rate
- probabilistic model
- document retrieval
- sound source
- information retrieval
- query expansion
- language modelling
- speech synthesis
- test collection
- retrieval model
- linear prediction
- vocal tract
- machine learning
- statistical language models
- background noise
- smoothing methods
- handwriting recognition
- broadcast news
- document ranking
- prior knowledge
- okapi bm
- language models for information retrieval
- additive noise
- image restoration
- maximum likelihood
- high dimensional
- image processing