Repository logo
 

Adapting an Unadaptable ASR System

Accepted version
Peer-reviewed

Type

Conference Object

Change log

Authors

Ma, R 
Qian, M 
Gales, MJF 
Knill, KM 

Abstract

As speech recognition model sizes and training data requirements grow, it is increasingly common for systems to only be available via APIs from online service providers rather than having direct access to models themselves. In this scenario it is challenging to adapt systems to a specific target domain. To address this problem we consider the recently released Open AI Whisper ASR as an example of a large-scale ASR system as to assess adaptation methods. An error correction based approach is adopted, as this does not require access to the model, but can be trained from either 1-best or N-best outputs that are normally available via the ASR API. LibriSpeech is used as the primary target domain for adaptation. The generalization ability of the system in two distinct dimensions are then evaluated. First, whether the form of correction model is portable to other speech recognition domains, and secondly whether it can be used for ASR models having a different architecture.

Description

Keywords

Journal Title

Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

Conference Name

INTERSPEECH 2023

Journal ISSN

2308-457X
1990-9772

Volume Title

Publisher

ISCA