Please use this identifier to cite or link to this item:
https://oxfordhealth-nhs.archive.knowledgearc.net/handle/123456789/333
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | External author(s) only | |
dc.date.accessioned | 2019-09-25T10:28:35Z | |
dc.date.available | 2019-09-25T10:28:35Z | |
dc.date.issued | 2019-09 | |
dc.identifier.citation | Bo Wang, Maria Liakata, Hao Ni, Terry Lyons, Alejo J Nevado-Holgado, Kate Saunders. A Path Signature Approach for Speech Emotion Recognition. INTERSPEECH 2019 September 15–19, 2019 | en |
dc.identifier.issn | 1990-9772 | |
dc.identifier.uri | https://oxfordhealth-nhs.archive.knowledgearc.net/handle/123456789/333 | |
dc.description.abstract | Automatic speech emotion recognition (SER) remains a difficult task within human-computer interaction, despite increasing interest in the research community. One key challenge is how to effectively integrate short-term characterisation of speech segments with long-term information such as temporal variations. Motivated by the numerical approximation theory of stochastic differential equations (SDEs), we propose the novel use of path signatures. The latter provide a pathwise definition to solve SDEs, for the integration of short speech frames. Furthermore we propose a hierarchical tree structure of path signatures, to capture both global and local information. A simple tree-based convolutional neural network (TBCNN) is used for learning the structural information stemming from dyadic path-tree signatures. Our experimental results on a widely used benchmark dataset demonstrate comparable performance to complex neural network based systems. Index Terms: speech emotion recognition, path signature feature, convolutional neural network | en |
dc.description.sponsorship | Supported by the NIHR | en |
dc.description.uri | http://dx.doi.org/10.21437/Interspeech.2019-2624 | |
dc.language.iso | en | en |
dc.subject | Emotions | en |
dc.subject | Mental Disorders | en |
dc.subject | Speech Emotion Recognition | en |
dc.title | A Path Signature Approach for Speech Emotion Recognition | en |
dc.type | Article | en |
Appears in Collections: | Digital Medicine |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.