Learning from past mistakes: improving automatic speech recognition output via noisy-clean phrase context modeling
By Prashanth Gurunath Shivakumar, Haoqi Li, Kevin Knight, Panayiotis G. Georgiou. 2018
29 citations recorded by the source corpus.
Abstract Automatic speech recognition (ASR) systems often make unrecoverable errors due to subsystem pruning (acoustic, language and pronunciation models); for example, pruning words due to acoustics using short-term context, prior to rescoring with long-term context based on linguistics. In this work, we model ASR as a phrase-based noisy transformation channel and propose an error correction system that can learn from the aggregate errors of all the independent modules constituting the ASR and attempt to invert those. The proposed system can exploit long-term context using a neural network language model and can better choose between existing ASR output possibilities as well as re-introduce previously pruned or unseen (Out-Of-Vocabulary) phrases. It provides corrections under poorly performing ASR conditions without degrading any accurate transcriptions; such corrections are greater on top of out-of-domain and mismatched data ASR. Our system consistently provides improvements over the baseline ASR, even when baseline is further optimized through Recurrent Neural Network (RNN) language model rescoring. This demonstrates that any ASR improvements can be exploited independently and that our proposed system can potentially still provide benefits on highly optimized ASR. Finally, we present an extensive analysis of the type of errors corrected by our system.
Subfields: Artificial Intelligence
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Ugur Kursuncu
- Manas Gaur
- Amit P. Sheth
- Eric Fosler-Lussier
- Prashant Serai
- Vishal Sunder
- Li Deng
- Dong Yu
Citations in the corpus
- Knowledge Infused Learning (K-IL): Towards Deep Incorporation of Knowledge in Deep Learning (2019, background): “Knowledge Infusion Layer In a many-to-one NLM (Shivakumar et al. 2018) network with T hidden layers, the T layer contains the learned representation before the output layer.”
- Hallucination of Speech Recognition Errors With Sequence to Sequence Learning (2021, methodology): “[13] explored modeling confusability at the level of phrases of words, and improved WER in a ASR system with a hybrid DNN-HMM acoustic model.”