Toward ALISP: A proposal for Automatic Language Independent Speech Processing

Chollet, Gérard; Černocký, Jaň; Constantinescu, Andreï; Deligne, Sabine; Bimbot, Frédéric

doi:10.1007/978-3-642-60087-6_32

Cited by 22 publications

(16 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our proposed solution involves the compression of speech containing biometric identifiers to a symbolic representation [1] that anonymizes the users' identity, and then on the other hand to use searchable symmetric encryption [2] to enable the finding of strings of symbols (e.g. phones) in an encrypted speech transcription.…”

Section: Proposed Architecturementioning

confidence: 99%

See 1 more Smart Citation

Privacy preserving encrypted phonetic search of speech data

Glackin¹,

Chollet²,

Dugan³

et al. 2017

2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

This is the accepted version of the paper.This version of the publication may differ from the final published version. Permanent ABSTRACTThis paper presents a strategy for enabling speech recognition to be performed in the cloud whilst preserving the privacy of users. The approach advocates a demarcation of responsibilities between the client and server-side components for performing the speech recognition task. On the client-side resides the acoustic model, which symbolically encodes the audio and encrypts the data before uploading to the server. The server-side then employs searchable encryption to enable the phonetic search of the speech content. Some preliminary results for speech encoding and searchable encryption are presented.

show abstract

Section: Proposed Architecturementioning

confidence: 99%

“…By reformulating the typical speech recognition task in such a way as to facilitate cloud computation, for example by reducing speech recognition to a search procedure [1], we demonstrate how secure speech processing in the cloud can be realized.…”

Section: Introductionmentioning

confidence: 99%

Privacy preserving encrypted phonetic search of speech data

Glackin¹,

Chollet²,

Dugan³

et al. 2017

2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…This is the originality of the ALISP approach. ALISP is a data-driven technique that was first developed for very low bit-rate speech coding [10], and then successfully adapted for other tasks such as speaker [14] and language recognition [11]. Schwarz [23] showed that data-driven approaches are also applicable for music signals.…”

Section: Introductionmentioning

confidence: 99%

Automatic detection of known advertisements in radio broadcast with data-driven ALISP transcriptions

Khemiri

Chollet

Petrovska‐Delacrétaz

2011

Multimed Tools Appl

Self Cite

View full text Add to dashboard Cite

This paper presents an audio indexing system to search for known advertisements in radio broadcast streams, using automatically acquired segmental units. These units, called Automatic Language Independent Speech Processing (ALISP) units, are acquired using temporal decomposition and vector quantization and modeled by Hidden Markov Models (HMMs). To detect commercials, ALISP transcriptions of reference advertisements are compared to the transcriptions of the test radio stream using the Levenshtein distance. The system is described and evaluated on one day broadcast audio streams from 11 French radio stations containing 2070 advertisements. With a set of 2,172 reference advertisements we achieve a mean precision rate of 99% with the corresponding recall value of 96%. Moreover, this system allowed us to detect some annotation errors.

show abstract

“…This database generation process is a major difficulty in adapting the systems to new tasks and languages. To avoid this problem, we propose to use automatic language independent speech processing (ALISP) tools [7] that provide a segmentation of the speech requiring neither phonetic nor orthographic transcriptions of the speech data.…”

Section: Introductionmentioning

confidence: 99%