Synthetic Speech Detection Using Deep Neural Networks

Simple item page
dc.contributor.advisorTzerpos, Vassilios
dc.contributor.authorReimao, Ricardo Amaral Martins
dc.date.accessioned2019-11-22T18:41:56Z
dc.date.available2019-11-22T18:41:56Z
dc.date.copyright2019-05
dc.date.issued2019-11-22
dc.date.updated2019-11-22T18:41:56Z
dc.degree.disciplineComputer Science
dc.degree.levelMaster's
dc.degree.nameMSc - Master of Science
dc.description.abstractWith the advancements in deep learning and other techniques, synthetic speech is getting closer to a natural sounding voice. Some of the state-of-art technologies achieve such a high level of naturalness that even humans have difficulties distinguishing real speech from computer generated speech. Moreover, these technologies allow a person to train a speech synthesizer with a target voice, creating a model that is able to reproduce someone's voice with high fidelity. With this research, we thoroughly analyze how synthetic speech is generated and propose deep learning methodologies to detect such synthesized utterances. We first collected a significant amount of real and synthetic utterances to create the Fake or Real (FoR) dataset. Then, we analyzed the performance of the latest deep learning models in the classification of such utterances. Our proposed model achieves 99.86% accuracy in synthetic speech detection, which is a significant improvement from a human performance (65.7%).
dc.identifier.urihttp://hdl.handle.net/10315/36698
dc.languageen
dc.rightsAuthor owns copyright, except where explicitly noted. Please contact the author directly with licensing requests.
dc.subjectArtificial intelligence
dc.subject.keywordsmachine learning
dc.subject.keywordssynthetic speech
dc.subject.keywordsdeep learning
dc.subject.keywordsartificial intelligence
dc.subject.keywordsdeep neural networks
dc.subject.keywordsneural networks
dc.subject.keywordssynthetic speech detection
dc.subject.keywordsTTS
dc.subject.keywordstext to speech
dc.subject.keywordsspeech generation
dc.subject.keywordsspeech synthesis
dc.subject.keywordsaudio classification
dc.subject.keywordsspeech classification
dc.titleSynthetic Speech Detection Using Deep Neural Networks
dc.typeElectronic Thesis or Dissertation

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Reimao_Ricardo_AM_2019_Masters.pdf
Size:
3.12 MB
Format:
Adobe Portable Document Format
Description:
Download
License bundle
Now showing 1 - 2 of 2
No Thumbnail Available
Name:
license.txt
Size:
1.87 KB
Format:
Plain Text
Description:
Download
No Thumbnail Available
Name:
YorkU_ETDlicense.txt
Size:
3.39 KB
Format:
Plain Text
Description:
Download

Collections

Computer Science and Engineering

All items in the YorkSpace institutional repository are protected by copyright, with all rights reserved except where explicitly noted.