DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Customization of Automatic Speech Recognition Engines for Rare Word Detection Without Costly Model Re-Training

Bhattacharjee, Mrinmoy and Motlicek, Petr and Nigmatulina, Iuliia and Helmke, Hartmut and Ohneiser, Oliver and Kleinert, Matthias and Ehr, Heiko (2023) Customization of Automatic Speech Recognition Engines for Rare Word Detection Without Costly Model Re-Training. In: 13th SESAR Innovation Days. 13th SESAR Innovation Days, 2023-11-27 - 2023-11-30, Sevilla, Spanien. doi: 10.61009/SID.2023.1.10.

[img] PDF


Thanks to Alexa, Siri or Google Assistant automatic speech recognition (ASR) has changed our daily life during the last decade. Prototypic applications in the air traffic management (ATM) domain are available. Recently pre-filling radar label entries by ASR support has reached the technology readiness level before industrialization (TRL6). However, seldom spoken and airspace related words relevant in the ATM context remain a challenge for sophisticated applications. Open-source ASR toolkits or large pre-trained models for experts - allowing to tailor ASR to new domains - can be exploited with a typical constraint on availability of certain amount of domain specific training data, i.e., typically transcribed speech for adapting acoustic and/or language models. In general, it is sufficient for a "universal" ASR engine to reliably recognize a few hundred words that form the vocabulary of the voice communications between air traffic controllers and pilots. However, for each airport some hundred dependent words that are seldom spoken need to be integrated. These challenging word entities comprise special airline designators and waypoint names like "dexon" or "burok", which only appear in a specific region. When used, they are highly informative and thus require high recognition accuracies. Allowing plug and play customization with a minimum expert manipulation assumes that no additional training is required, i.e., fine-tuning the universal ASR. This paper presents an innovative approach to automatically integrate new specific word entities to the universal ASR system. The recognition rate of these region-specific word entities with respect to the universal ASR increases by a factor of 6.

Item URL in elib:https://elib.dlr.de/199971/
Document Type:Conference or Workshop Item (Speech)
Title:Customization of Automatic Speech Recognition Engines for Rare Word Detection Without Costly Model Re-Training
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Nigmatulina, IuliiaIdiap, University of ZurichUNSPECIFIEDUNSPECIFIED
Helmke, HartmutUNSPECIFIEDhttps://orcid.org/0000-0002-1939-0200UNSPECIFIED
Ohneiser, OliverUNSPECIFIEDhttps://orcid.org/0000-0002-5411-691XUNSPECIFIED
Kleinert, MatthiasUNSPECIFIEDhttps://orcid.org/0000-0002-0782-4147UNSPECIFIED
Date:28 November 2023
Journal or Publication Title:13th SESAR Innovation Days
Refereed publication:Yes
Open Access:Yes
Gold Open Access:No
In ISI Web of Science:No
Keywords:Speech Recognition; Model Adaptation; Integration of prior knowledge; Customization of model, Rare-word integration
Event Title:13th SESAR Innovation Days
Event Location:Sevilla, Spanien
Event Type:international Conference
Event Start Date:27 November 2023
Event End Date:30 November 2023
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Aeronautics
HGF - Program Themes:Air Transportation and Impact
DLR - Research area:Aeronautics
DLR - Program:L AI - Air Transportation and Impact
DLR - Research theme (Project):L - Integrated Flight Guidance
Location: Braunschweig
Institutes and Institutions:Institute of Flight Guidance > Controller Assistance
Deposited By: Ohneiser, Oliver
Deposited On:29 Nov 2023 07:34
Last Modified:24 Apr 2024 21:00

Repository Staff Only: item control page

Help & Contact
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.