Quality Evaluation of Reverberation in Audioband Speech Signals

Lima, Amaro A. de; Netto, Sérgio L.; Biscainho, Luiz W. P.; Freeland, Fabio P.; Bispo, Bruno; Jesus, Rafael A. de; Schafer, Ronald W.; Said, Amir; Lee, Bowon; Kalker, Ton

doi:10.1007/978-3-642-05197-5_28

Cited by 10 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…where N is the number of microphone elements in the array. Both noise and reverberance degrade the intelligibility and quality of speech [38], [39]. The ultimate goal of speech enhancement is therefore to recover the speech signal s from the single or multi-channel noisy signal ŷ.…”

Section: A Problem Formulationmentioning

confidence: 99%

On the Predictive Power of Objective Intelligibility Metrics for the Subjective Performance of Deep Complex Convolutional Recurrent Speech Enhancement Networks

Gelderblom¹,

Tronstad²,

Svendsen³

et al. 2022

Preprint

View full text Add to dashboard Cite

<div>Speech enhancement (SE) systems aim to improve the quality and intelligibility of degraded speech signals obtained from far-field microphones. Subjective evaluation of the intelligibility performance of these SE systems is uncommon. Instead, objective intelligibility measures (OIMs) are generally used to predict subjective performance increases. Many recent deep learning based SE systems, are expected to improve the intelligibility of degraded speech as measured by OIMs. </div><div><br></div><div>However, validation of the OIMs for this purpose is lacking. Therefore, in this study, we evaluate the predictive performance of five popular OIMs. We compare the metrics' predictions with subjective results. For this purpose, we recruited 50 human listeners, and subjectively tested both single channel and multi-channel Deep Complex Convolutional Recurrent Network (DCCRN) based speech systems. </div><div><br></div><div>We find that none of the OIMs gave reliable predictions, and that all OIMs overestimated the intelligibility of `enhanced' speech signals. </div>

show abstract

Section: A Problem Formulationmentioning

confidence: 99%

On the Predictive Power of Objective Intelligibility Metrics for the Subjective Performance of Deep Complex Convolutional Recurrent Speech Enhancement Networks

Gelderblom¹,

Tronstad²,

Svendsen³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The W-PESQ achieves a correlation of 80% with MOS when assessing speech impairment by reverberation although it was not designed for this purpose [25], [26]. Hence, the W-PESQ was applied to perceptually evaluate the performance of the proposed controller for attenuating the occlusion effect.…”

Section: Simulation Configurationsmentioning

confidence: 99%

A Cepstral Method to Estimate the Stable Optimal Solution for Feedforward Occlusion Cancellation in Hearing Aids

Bispo

Borges

2020

JCIS

View full text Add to dashboard Cite

The occlusion effect is a common complaint from users of hearing aids with narrow or unvented earmolds. This phenomenon makes the user hear his own voice muffled. In the scientific literature, fixed and adaptive controllers have been proposed for occlusion effect reduction. This work proposes a cepstral method to estimate the stable optimal solution for feedforward occlusion cancellation and a fixed controller that utilizes this estimate to reduce the occlusion effect in hearing aids. The cepstral method operates on a feedback structure at a calibration process. Simulations have shown that the performance of the cepstral method improves as the length of the signal uttered by the hearing aid user increases, resulting in estimates with average normalized misalignment less than -19 and -34 dB for signals lasting 1.5 and 5 s. The estimates are significantly more accurate below 500 Hz, which is the frequency range common to the occlusion effect. In addition, results have pointed out that the controller attenuated the occlusion effect, averagely decreasing by 0.17 dB the distortion power and increasing by 0.13 the objective perceptual quality MOS-LQO score.

show abstract

“…O W-PESQ possui uma correlação de 80% com MOS ao avaliar o comprometimento da fala por reverberação embora não tenha sido projetado para esse propósito [13], [14]. Não há estudos que comprovem que a oclusão, queé caracterizada pela amplificação de faixas estreitas em baixas frequências, e adequadamente quantificada por esse algoritmo.…”

Section: Sistema De Oclusãounclassified

Método cepstral para estimação da resposta ao impulso do efeito de oclusão em sistemas de auxílio à audição

Bispo¹,

Borges²,

Costa³

2019

Anais De XXXVII Simpósio Brasileiro De Telecomunicações E Processamento De Sinais

View full text Add to dashboard Cite

Resumo-O efeito de oclusãoé uma queixa comum de usuários de aparelhos auditivos que utilizam moldes sem ventilação ou com ventilação estreita, os quais promovem uma sensação de abafamento da própria voz. Na literatura, controladores fixos e adaptativos têm sido propostos para a redução desse efeito, o qual possui uma soluçãoótima não-causal. Este trabalho propõe um método cepstral para identificar a resposta ao impulso da oclusão e um sistema de controle fixo que utiliza essa estimativa para redução do efeito de oclusão em sistemas de auxílioà audição. Simulações indicam que a eficiência do método cepstral depende do tamanho da janela de análise, resultando em estimativas com desalinhamento média de-28,77 dB para janelas de análise de 2 s. E que o controlador aproxima o sinal a ser ouvido pelo usuário do seu valor ideal, diminuindo em 0,34 dB a potência do erro e aumentando em 0,12 MOS-LQO a qualidade perceptual do sinal. Palavras-Chave-Efeito de oclusão, aparelho auditivo, análise cepstral, controle ativo.

show abstract

Quality Evaluation of Reverberation in Audioband Speech Signals

Cited by 10 publications

References 9 publications

On the Predictive Power of Objective Intelligibility Metrics for the Subjective Performance of Deep Complex Convolutional Recurrent Speech Enhancement Networks

On the Predictive Power of Objective Intelligibility Metrics for the Subjective Performance of Deep Complex Convolutional Recurrent Speech Enhancement Networks

A Cepstral Method to Estimate the Stable Optimal Solution for Feedforward Occlusion Cancellation in Hearing Aids

Método cepstral para estimação da resposta ao impulso do efeito de oclusão em sistemas de auxílio à audição

Contact Info

Product

Resources

About