Ask2Mask: Guided Data Selection for Masked Speech Modeling
Identifikátory výsledku
Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F22%3APU147442" target="_blank" >RIV/00216305:26230/22:PU147442 - isvavai.cz</a>
Výsledek na webu
<a href="https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9806175" target="_blank" >https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9806175</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1109/JSTSP.2022.3186162" target="_blank" >10.1109/JSTSP.2022.3186162</a>
Alternativní jazyky
Jazyk výsledku
angličtina
Název v původním jazyce
Ask2Mask: Guided Data Selection for Masked Speech Modeling
Popis výsledku v původním jazyce
Masked speech modeling (MSM) methods such as wav2vec2 or w2v-BERT learn representations over speech frames which are randomlymaskedwithin an utterance. While thesemethods improve performance of Automatic Speech Recognition (ASR) systems, they have one major limitation. They treat all unsupervised speech samples with equal weight, which hinders learning as not all samples have relevant information to learn meaningful representations. In this work, we address this limitation. We propose ask2mask (ATM), a novel approach to focus on specific samples during MSM pre-training. ATM employs an external ASR model or scorer to weight unsupervised input samples in two different ways: 1) A fine-grained data selection is performed by masking over the highly confident input frames as chosen by the scorer. This allows themodel to learnmeaningful representations. 2) ATM is further extended to focus at utterance-level by weighting the final MSM loss with the utterance-level confidence score. We conduct fine-tuning experiments on two well-benchmarked corpora: LibriSpeech (matching the pre-training data) and Commonvoice, TED-LIUM, AMI and CHiME-6 (not matching the pre-training data). The results substantiate the efficacy of ATM on significantly improving the recognition performance under mismatched conditions (up to 11.6% relative over published results and upto 4.46% relative over our internal baseline) while still yielding modest improvements under matched conditions.
Název v anglickém jazyce
Ask2Mask: Guided Data Selection for Masked Speech Modeling
Popis výsledku anglicky
Masked speech modeling (MSM) methods such as wav2vec2 or w2v-BERT learn representations over speech frames which are randomlymaskedwithin an utterance. While thesemethods improve performance of Automatic Speech Recognition (ASR) systems, they have one major limitation. They treat all unsupervised speech samples with equal weight, which hinders learning as not all samples have relevant information to learn meaningful representations. In this work, we address this limitation. We propose ask2mask (ATM), a novel approach to focus on specific samples during MSM pre-training. ATM employs an external ASR model or scorer to weight unsupervised input samples in two different ways: 1) A fine-grained data selection is performed by masking over the highly confident input frames as chosen by the scorer. This allows themodel to learnmeaningful representations. 2) ATM is further extended to focus at utterance-level by weighting the final MSM loss with the utterance-level confidence score. We conduct fine-tuning experiments on two well-benchmarked corpora: LibriSpeech (matching the pre-training data) and Commonvoice, TED-LIUM, AMI and CHiME-6 (not matching the pre-training data). The results substantiate the efficacy of ATM on significantly improving the recognition performance under mismatched conditions (up to 11.6% relative over published results and upto 4.46% relative over our internal baseline) while still yielding modest improvements under matched conditions.
Klasifikace
Druh
J<sub>imp</sub> - Článek v periodiku v databázi Web of Science
CEP obor
—
OECD FORD obor
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Návaznosti výsledku
Projekt
—
Návaznosti
R - Projekt Ramcoveho programu EK
Ostatní
Rok uplatnění
2022
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Údaje specifické pro druh výsledku
Název periodika
IEEE J-STSP
ISSN
1932-4553
e-ISSN
1941-0484
Svazek periodika
16
Číslo periodika v rámci svazku
6
Stát vydavatele periodika
US - Spojené státy americké
Počet stran výsledku
10
Strana od-do
1357-1366
Kód UT WoS článku
000870301500019
EID výsledku v databázi Scopus
2-s2.0-85133786585