Multitask learning in connectionist robust ASR using recurrent neural networks

Parveen, Shahla; Green, Phil

doi:10.21437/eurospeech.2003-500

Cited by 20 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…O Reconhecimento Automático de Gênero (Automatic Gender Recognition -AGR) referese à identificac ¸ão de gênero de um locutor a partir de um sinal de áudio [Kabil et al 2018]. Sistemas de AGR são utilizados, por exemplo, para a melhora da recomendac ¸ão de produtos em estratégias de marketing [Shepstone et al 2013]; análise de voz em cenários de investigac ¸ão de crimes [Nair and Savithri 2021]; melhora de sistemas interac ¸ão humanocomputador [La Mura and Lamberti 2020] e no auxílio de reconhecimento automático de voz [Parveen and Green 2003].…”

Section: Introduc ¸ãOunclassified

A Comparison of Deep Learning Architectures for Automatic Gender Recognition from Audio Signals

Ferreira

Oliveira

Silva

et al. 2021

ENIAC

View full text Add to dashboard Cite

O reconhecimento de gênero a partir da fala é um problema relacionado à análise de fala humana, e possui diversas aplicações que vão desde a personalização na recomendação de produtos à ciência forense. A identificação da eficiência e custos de diferentes abordagens que lidam com esse problema é imprescindível. Este trabalho tem como foco investigar e comparar a eficiência e custos de diferentes arquiteturas de deep learning para o reconhecimento de gênero a partir da fala. Os resultados mostram que o modelo convolucional unidimensional consegue os melhores resultados. No entanto, constatou-se que o modelo fully connected apresentou resultados próximos com menor custo, tanto no uso de memória, quanto no tempo de treinamento.

show abstract

Section: Introduc ¸ãOunclassified

A Comparison of Deep Learning Architectures for Automatic Gender Recognition from Audio Signals

Ferreira

Oliveira

Silva

et al. 2021

ENIAC

View full text Add to dashboard Cite

show abstract

“…The technique of multi-task learning has been successfully employed in a wide range of areas over the last few decades. Under the area of automatic speech recognition, MTL with speech enhancement [17], and different target prediction [18] have been explored to improve model performance. In speaker verification, MTL with word prediction [19], phonetic prediction [20], and speaker attribute prediction [21] have been used to learn robust speaker representation for superior model performance.…”

Section: Introductionmentioning

confidence: 99%

Review and update of the Hong Kong Epilepsy Guideline on status epilepticus

Fung

2017

Hong Kong Med J

View full text Add to dashboard Cite

Due to the scarcity of publicly available diarization data, the model performance can be improved by training a single model with data from different domains. In this work, we propose to incorporate domain information to train a single end-to-end diarization model for multiple domains. First, we employ domain adaptive training with parameter-efficient adapters for on-the-fly model reconfiguration. Second, we introduce an auxiliary domain classification task to make the diarization model more domain-aware. For seen domains, the combination of our proposed methods reduces the absolute DER from 17.66% to 16.59% when compared with the baseline. During inference, adapters from ground-truth domains are not available for unseen domains. We demonstrate our model exhibits a stronger generalizability to unseen domains when adapters are removed. For two unseen domains, this improves the DER performance from 39.91% to 23.09% and 25.32% to 18.76% over the baseline, respectively.

show abstract

Improving Articulatory Feature and Phoneme Recognition Using Multitask Learning

Rasipuram

Magimai-Doss

2011

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Multitask learning in connectionist robust ASR using recurrent neural networks

Cited by 20 publications

References 12 publications

A Comparison of Deep Learning Architectures for Automatic Gender Recognition from Audio Signals

A Comparison of Deep Learning Architectures for Automatic Gender Recognition from Audio Signals

Review and update of the Hong Kong Epilepsy Guideline on status epilepticus

Improving Articulatory Feature and Phoneme Recognition Using Multitask Learning

Contact Info

Product

Resources

About