Noise-robust speech triage

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F18%3APU127787" target="_blank" >RIV/00216305:26230/18:PU127787 - isvavai.cz</a>

Výsledek na webu

<a href="https://asa.scitation.org/doi/10.1121/1.5031029" target="_blank" >https://asa.scitation.org/doi/10.1121/1.5031029</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1121/1.5031029" target="_blank" >10.1121/1.5031029</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Noise-robust speech triage

Popis výsledku v původním jazyce

A method is presented in which conventional speech algorithms are applied, with no modifications, to improve their performance in extremely noisy environments. It has been demonstrated that, for eigen-channel algorithms, pre-training multiple speaker identification (SID) models at a lattice of signal-to-noise-ratio (SNR) levels and then performing SID using the appropriate SNR dependent model was successful in mitigating noise at all SNR levels. In those tests, it was found that SID performance was optimized when the SNR of the testing and training data were close or identical. In this current effort multiple i-vector algorithms were used, greatly improving both processing throughput and equal error rate classification accuracy. Using identical approaches in the same noisy environment, performance of SID, language identification, gender identification, and diarization were significantly improved. A critical factor in this improvement is speech activity detection (SAD) that performs reliably in extremely noisy environments, where the speech itself is barely audible. To optimize SAD operation at all SNR levels, two algorithms were employed. The first maximized detection probability at low levels (10 dB SNR < 10 dB) using just the voiced speech envelope, and the second exploited features extracted from the original speech to improve overall accuracy at higher quality levels (SNR10 dB).

Název v anglickém jazyce

Noise-robust speech triage

Popis výsledku anglicky

A method is presented in which conventional speech algorithms are applied, with no modifications, to improve their performance in extremely noisy environments. It has been demonstrated that, for eigen-channel algorithms, pre-training multiple speaker identification (SID) models at a lattice of signal-to-noise-ratio (SNR) levels and then performing SID using the appropriate SNR dependent model was successful in mitigating noise at all SNR levels. In those tests, it was found that SID performance was optimized when the SNR of the testing and training data were close or identical. In this current effort multiple i-vector algorithms were used, greatly improving both processing throughput and equal error rate classification accuracy. Using identical approaches in the same noisy environment, performance of SID, language identification, gender identification, and diarization were significantly improved. A critical factor in this improvement is speech activity detection (SAD) that performs reliably in extremely noisy environments, where the speech itself is barely audible. To optimize SAD operation at all SNR levels, two algorithms were employed. The first maximized detection probability at low levels (10 dB SNR < 10 dB) using just the voiced speech envelope, and the second exploited features extracted from the original speech to improve overall accuracy at higher quality levels (SNR10 dB).

Druh

J_imp - 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)

Projekt

<a href="/cs/project/VI20152020025" target="_blank" >VI20152020025: Dolování infoRmAcí z řeči Pořízené vzdÁlenými miKrofony - DRAPÁK</a><br>

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2018

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Journal of the Acoustical Society of America

ISSN

0001-4966

e-ISSN

1520-8524

Svazek periodika

143

Číslo periodika v rámci svazku

4

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

8

Strana od-do

2313-2320

Kód UT WoS článku

000430570900039

EID výsledku v databázi Scopus

2-s2.0-85045888415