Cornell University
EchoNose - AN ACOUSTIC NASAL INTERFACE FOR SILENT SPEECH RECOGNITION
Abstract
dc:description.abstractEchoNose is a minimally-invasive wearable device capable of detecting silent speech words and articulators (such as tongue movement) using active acoustic sensing through the nasal cavity. This effect is possible due to the interconnectivity of the upper respiratory tract cavities. We explore the feasibility of EchoNose by (1) continuously tracking speech articulators using an electromagnetic articulograph (EMA) for ground truth, and (2) conducting a silent speech user study with 6 participants comparing EchoNose against CV performance on a list of 21 common words and 10 “tongue-specific” words.
Degree
thesis:*- Name thesis:degree_name
- M.S., Computer Science
- Level thesis:degree_level
- Master of Science
- Discipline thesis:degree_discipline
- Computer Science
- Grantor
- Cornell University
- Year dc:date.issued
- 2022
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Steeper, Benjamin
- Chair dc:contributor.chair
-
- Zhang, Cheng
- Committee member dc:contributor.committeemember
-
- Tilsen, Sam
Subjects
dc:subject × 6Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Dc Identifier Bibid
- 15578970
- Dc Identifier Other
-
ProQuest Submission ID: 11547
ProQuest Publication ID: 29323765