Year: 2023
Type: conference-paper
Abstract: We present a first large-scale public synthetic dataset for contextual spellchecking customization of automatic speech recognition (ASR) with focus on diverse rare and out-of-vocabulary (OOV) phrases, such as proper names or terms. The proposed approach allows creating millions of realistic examples of corrupted ASR hypotheses and simulate non-trivial biasing lists for the customization task. Furthermore, we propose injecting two types of “hard negatives” to the simulated biasing lists in training examples and describe our procedures to automatically mine them. We report experiments with training an open-source customization model on the proposed dataset and show that the injection of hard negative biasing phrases decreases WER and the number of false alarms.
Author: Alexandra Antonova
Institution: Moscow Institute of Physics and Technology
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4391021725