Evaluating Near End Listening Enhancement Algorithms in Realistic Environments

Carol Chermaz; Cassia Valentini Botinhao; Henning Schepker; Simon King

doi:10.21437/Interspeech.2019-1800

Evaluating Near End Listening Enhancement Algorithms in Realistic Environments

Carol Chermaz, Cassia Valentini Botinhao, Henning Schepker, Simon King

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract / Description of output

Speech playback (e.g., TV, radio, public address) becomes harder to understand in the presence of noise and reverberation. NELE (Near End Listening Enhancement) algorithms can improve intelligibility by modifying the signal before it is played back. Substantial intelligibility improvements have been achieved in the lab for both natural and synthetic speech. However, evidence is still scarce on how these algorithms work under conditions of realistic noise and reverberation.

We present a realistic test platform, featuring two representative everyday scenarios in which speech playback may occur (in the presence of both noise and reverberation): a domestic space (living room) and a public space (cafeteria). The generated stimuli are evaluated by measuring keyword accuracy rates in a listening test with normal hearing subjects.

We use the new platform to compare three state-of-theart NELE algorithms, employing either noise-adaptive or nonadaptive strategies, and with or without compensation for reverberation.

Original language	English
Title of host publication	Proceedings Interspeech 2019
Publisher	International Speech Communication Association
Pages	1373-1377
Number of pages	5
DOIs	https://doi.org/10.21437/Interspeech.2019-1800
Publication status	Published - 19 Sept 2019
Event	Interspeech 2019 - Graz, Austria Duration: 15 Sept 2019 → 19 Sept 2019 https://www.interspeech2019.org/

Publication series

Name
Publisher	International Speech Communication Association
ISSN (Electronic)	1990-9772

Conference

Conference	Interspeech 2019
Country/Territory	Austria
City	Graz
Period	15/09/19 → 19/09/19
Internet address	https://www.interspeech2019.org/

Keywords / Materials (for Non-textual outputs)

NELE
Near End Listening Enhancement
realistic noise
reverberation
speech modifications

Access to Document

10.21437/Interspeech.2019-1800Licence: All Rights Reserved

evaluating_near_end_listening_CHERMAZ_DoA170619_AFVAccepted author manuscript, 1.84 MB

https://www.isca-speech.org/archive/Interspeech_2019/abstracts/1800.htmlLicence: All Rights Reserved

Cite this

@inproceedings{223027eca4b74768944c9f53fc0645b4,

title = "Evaluating Near End Listening Enhancement Algorithms in Realistic Environments",

abstract = "Speech playback (e.g., TV, radio, public address) becomes harder to understand in the presence of noise and reverberation. NELE (Near End Listening Enhancement) algorithms can improve intelligibility by modifying the signal before it is played back. Substantial intelligibility improvements have been achieved in the lab for both natural and synthetic speech. However, evidence is still scarce on how these algorithms work under conditions of realistic noise and reverberation.We present a realistic test platform, featuring two representative everyday scenarios in which speech playback may occur (in the presence of both noise and reverberation): a domestic space (living room) and a public space (cafeteria). The generated stimuli are evaluated by measuring keyword accuracy rates in a listening test with normal hearing subjects.We use the new platform to compare three state-of-theart NELE algorithms, employing either noise-adaptive or nonadaptive strategies, and with or without compensation for reverberation.",

keywords = "NELE, Near End Listening Enhancement, realistic noise, reverberation, speech modifications",

author = "Carol Chermaz and {Valentini Botinhao}, Cassia and Henning Schepker and Simon King",

year = "2019",

month = sep,

day = "19",

doi = "10.21437/Interspeech.2019-1800",

language = "English",

publisher = "International Speech Communication Association",

pages = "1373--1377",

booktitle = "Proceedings Interspeech 2019",

note = "Interspeech 2019 ; Conference date: 15-09-2019 Through 19-09-2019",

url = "https://www.interspeech2019.org/",

}

TY - GEN

T1 - Evaluating Near End Listening Enhancement Algorithms in Realistic Environments

AU - Chermaz, Carol

AU - Valentini Botinhao, Cassia

AU - Schepker, Henning

AU - King, Simon

PY - 2019/9/19

Y1 - 2019/9/19

N2 - Speech playback (e.g., TV, radio, public address) becomes harder to understand in the presence of noise and reverberation. NELE (Near End Listening Enhancement) algorithms can improve intelligibility by modifying the signal before it is played back. Substantial intelligibility improvements have been achieved in the lab for both natural and synthetic speech. However, evidence is still scarce on how these algorithms work under conditions of realistic noise and reverberation.We present a realistic test platform, featuring two representative everyday scenarios in which speech playback may occur (in the presence of both noise and reverberation): a domestic space (living room) and a public space (cafeteria). The generated stimuli are evaluated by measuring keyword accuracy rates in a listening test with normal hearing subjects.We use the new platform to compare three state-of-theart NELE algorithms, employing either noise-adaptive or nonadaptive strategies, and with or without compensation for reverberation.

AB - Speech playback (e.g., TV, radio, public address) becomes harder to understand in the presence of noise and reverberation. NELE (Near End Listening Enhancement) algorithms can improve intelligibility by modifying the signal before it is played back. Substantial intelligibility improvements have been achieved in the lab for both natural and synthetic speech. However, evidence is still scarce on how these algorithms work under conditions of realistic noise and reverberation.We present a realistic test platform, featuring two representative everyday scenarios in which speech playback may occur (in the presence of both noise and reverberation): a domestic space (living room) and a public space (cafeteria). The generated stimuli are evaluated by measuring keyword accuracy rates in a listening test with normal hearing subjects.We use the new platform to compare three state-of-theart NELE algorithms, employing either noise-adaptive or nonadaptive strategies, and with or without compensation for reverberation.

KW - NELE

KW - Near End Listening Enhancement

KW - realistic noise

KW - reverberation

KW - speech modifications

UR - https://www.interspeech2019.org/calls/papers/

UR - https://www.isca-speech.org/iscaweb/index.php/publications

U2 - 10.21437/Interspeech.2019-1800

DO - 10.21437/Interspeech.2019-1800

M3 - Conference contribution

SP - 1373

EP - 1377

BT - Proceedings Interspeech 2019

PB - International Speech Communication Association

T2 - Interspeech 2019

Y2 - 15 September 2019 through 19 September 2019

ER -

Evaluating Near End Listening Enhancement Algorithms in Realistic Environments

Abstract / Description of output

Publication series

Conference

Keywords / Materials (for Non-textual outputs)

Access to Document

Other files and links

Fingerprint

Enriched communication across the lifespan

Best student paper of Interspeech 2019

Cite this

Evaluating Near End Listening Enhancement Algorithms in Realistic Environments

Abstract / Description of output

Publication series

Conference

Keywords / Materials (for Non-textual outputs)

Access to Document

Other files and links

Fingerprint

Projects

Enriched communication across the lifespan

Prizes

Best student paper of Interspeech 2019

Cite this