Word Order does not Matter for Speech Recognition
Vineel Pratap, Qiantong Xu, Tatiana Likhomanenko, Gabriel Synnaeve, Ronan Collobert
Abstract
In this paper, we study training of automatic speech recognition system in a weakly supervised setting where the order of words in transcript labels of the audio training data is not known. We train a word-level acoustic model which aggregates the distribution of all output frames using LogSumExp operation and uses a cross-entropy loss to match with the ground-truth words distribution. Using the pseudo-labels generated from this model on the training set, we then train a letter-based acoustic model using Connectionist Temporal Classification loss. Our system achieves 2.3%/4.6% on test-clean/test-other subsets of LibriSpeech, which closely matches with the supervised baseline's performance.
BibTeX
@inproceedings{icassp2022_wordorderdoesnot,
title = {Word Order does not Matter for Speech Recognition},
author = {Vineel Pratap and Qiantong Xu and Tatiana Likhomanenko and Gabriel Synnaeve and Ronan Collobert},
booktitle = {ICASSP 2022},
year = {2022}
}