Blind Vocoder Speech Reconstruction using Generative Adversarial Networks

Yoav Blum; David Burshtein

doi:https://doi.org/10.1109/WIFS47025.2019.9035106

Blind Vocoder Speech Reconstruction using Generative Adversarial Networks

Yoav Blum, David Burshtein

School of Electrical Engineering

Tel Aviv University

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

Abstract

The problem of reconstructing vocoder acoustic parameters using only encoded bit stream data is considered with applications to forensics and reverse engineering. Wasserstein generative adversarial networks (GANs) and CycleGANs, that map two unpaired domains, are used. It is shown that it is possible to reconstruct key acoustic parameters such as linear predictive coefficients (LPCs) when these parameters are encoded using scalar quantization. It is further shown that speech reconstruction is possible to some extent when it is known that the vocoder belongs to the family of code excited linear prediction (CELP) models, but the coded bit frame structure is unknown.

Original language	English
Title of host publication	2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781728132174
DOIs	https://doi.org/10.1109/WIFS47025.2019.9035106
State	Published - Dec 2019
Event	2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019 - Delft, Netherlands Duration: 9 Dec 2019 → 12 Dec 2019

Publication series

Name	2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019

Conference

Conference	2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019
Country/Territory	Netherlands
City	Delft
Period	9/12/19 → 12/12/19

All Science Journal Classification (ASJC) codes

Information Systems and Management
Safety, Risk, Reliability and Quality
Library and Information Sciences
Computer Networks and Communications

Access to Document

https://doi.org/10.1109/WIFS47025.2019.9035106

Cite this

Blum, Y., & Burshtein, D. (2019). Blind Vocoder Speech Reconstruction using Generative Adversarial Networks. In 2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019 Article 9035106 (2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/WIFS47025.2019.9035106

@inproceedings{4bd8a6aaa5574716869ee218d81941e3,

title = "Blind Vocoder Speech Reconstruction using Generative Adversarial Networks",

abstract = "The problem of reconstructing vocoder acoustic parameters using only encoded bit stream data is considered with applications to forensics and reverse engineering. Wasserstein generative adversarial networks (GANs) and CycleGANs, that map two unpaired domains, are used. It is shown that it is possible to reconstruct key acoustic parameters such as linear predictive coefficients (LPCs) when these parameters are encoded using scalar quantization. It is further shown that speech reconstruction is possible to some extent when it is known that the vocoder belongs to the family of code excited linear prediction (CELP) models, but the coded bit frame structure is unknown.",

author = "Yoav Blum and David Burshtein",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019 ; Conference date: 09-12-2019 Through 12-12-2019",

year = "2019",

month = dec,

doi = "https://doi.org/10.1109/WIFS47025.2019.9035106",

language = "الإنجليزيّة",

series = "2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019",

address = "الولايات المتّحدة",

}

Blum, Y & Burshtein, D 2019, Blind Vocoder Speech Reconstruction using Generative Adversarial Networks. in 2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019., 9035106, 2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019, Institute of Electrical and Electronics Engineers Inc., 2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019, Delft, Netherlands, 9/12/19. https://doi.org/10.1109/WIFS47025.2019.9035106

Blind Vocoder Speech Reconstruction using Generative Adversarial Networks. / Blum, Yoav; Burshtein, David.
2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019. Institute of Electrical and Electronics Engineers Inc., 2019. 9035106 (2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019).

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

TY - GEN

T1 - Blind Vocoder Speech Reconstruction using Generative Adversarial Networks

AU - Blum, Yoav

AU - Burshtein, David

PY - 2019/12

Y1 - 2019/12

N2 - The problem of reconstructing vocoder acoustic parameters using only encoded bit stream data is considered with applications to forensics and reverse engineering. Wasserstein generative adversarial networks (GANs) and CycleGANs, that map two unpaired domains, are used. It is shown that it is possible to reconstruct key acoustic parameters such as linear predictive coefficients (LPCs) when these parameters are encoded using scalar quantization. It is further shown that speech reconstruction is possible to some extent when it is known that the vocoder belongs to the family of code excited linear prediction (CELP) models, but the coded bit frame structure is unknown.

AB - The problem of reconstructing vocoder acoustic parameters using only encoded bit stream data is considered with applications to forensics and reverse engineering. Wasserstein generative adversarial networks (GANs) and CycleGANs, that map two unpaired domains, are used. It is shown that it is possible to reconstruct key acoustic parameters such as linear predictive coefficients (LPCs) when these parameters are encoded using scalar quantization. It is further shown that speech reconstruction is possible to some extent when it is known that the vocoder belongs to the family of code excited linear prediction (CELP) models, but the coded bit frame structure is unknown.

UR - http://www.scopus.com/inward/record.url?scp=85083040188&partnerID=8YFLogxK

U2 - https://doi.org/10.1109/WIFS47025.2019.9035106

DO - https://doi.org/10.1109/WIFS47025.2019.9035106

M3 - منشور من مؤتمر

T3 - 2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019

BT - 2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019

Y2 - 9 December 2019 through 12 December 2019

ER -

Blum Y, Burshtein D. Blind Vocoder Speech Reconstruction using Generative Adversarial Networks. In 2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019. Institute of Electrical and Electronics Engineers Inc. 2019. 9035106. (2019 IEEE International Workshop on Information Forensics and Security, WIFS 2019). doi: https://doi.org/10.1109/WIFS47025.2019.9035106

Blind Vocoder Speech Reconstruction using Generative Adversarial Networks

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this