Recognition of bimodal produced speech based on Support Vector Machines(Conference Paper)
[Prepoznavanje bimodalnog govora bazirano na metodi potpornih vektora]
Retrieving additional authors...
- aElektrotehnički Fakultet, Univerzitet U Banjoj Luci, Patre 5, Banja Luka, 78000, Bosnia and Herzegovina
- bElektrotehnički Fakultet, Univerzitet U Beogradu, Bulevar kralja Aleksandra 73, Beograd, 11120, Serbia
- cCentar Za Unapredenje Životnih Aktivnosti, Gospodar Jovanova 35, Beograd, 11000, Serbia
- dVisoka Škola Tehničkih Strukovnih Studija, Svetog Save 65, Čačak, 32000, Serbia
Retrieving additional affiliations...
Abstract
This paper presents the results of experiments in recognition of whispered speech, as a specific mode of verbal communication, using SVM (Support Vector Machines) classifier. For training and testing purposes, part of the speech database Whi-Spe with male speakers has been exploited. In matched scenarios average word recognition rate was 99.3% (for normally phonated speech) and 97.8% (for whisper). In mismatched scenarios, recognition of whisper with training on normal speech utterances was 75.4%, whereas recognition of whisper-trained normal speech was with recognition rate of 81.3%. © 2017 IEEE.
Indexed keywords
| Engineering controlled terms: | SpeechSpeech communicationSupport vector machines |
|---|---|
| Engineering uncontrolled terms | Govorna baza Whi-Spemetoda potpornih vektoraprepoznavanje govoraRecognition ratesSpeech utteranceTraining and testingVerbal communicationsWhispered speech |
| Engineering main heading: | Speech recognition |
- ISBN: 978-153863072-3
- Source Type: Conference Proceeding
- Original language: Bosnian
- DOI: 10.1109/TELFOR.2017.8249360
- Document Type: Conference Paper
- Sponsors: "Telekom SRBIJA" a.d.,Ericsson,et al.,ETF - School of Electrical Engineering, University of Belgrade,Ministry of Trade, Turism and Telecommunications,VLATACOM d.o.o
- Publisher: Institute of Electrical and Electronics Engineers Inc.
© Copyright 2018 Elsevier B.V., All rights reserved.
Cited by 0 documents
{"topic":{"name":"Speech Communication; Neural Network; Audio Signal Processing","id":61698,"uri":"Topic/61698","prominencePercentile":58.277027,"prominencePercentileString":"58.277","overallScholarlyOutput":0},"dig":"1524f452c86b1f77a831aac958f159b03eb0669088488551527c47152a8d03a8"}