Abstract
Personal Voice Assistants (PVAs) are used to interact with digital environments and computer systems using speech. In this work we describe how to identify the room in which the speaker is located. Only the audio signal is used for identification without using any other sensor input. We use the output of existing trained models for speaker identification in combination with a Support Vector Machine (SVM) to perform room identification. This method allows us to re-use existing elements of PVA eco-systems and an intensive training phase is not required. In our evaluation rooms can be identified with almost 90% accuracy. Room identification might be used as additional security mechanism and the work shows that speech signals recorded by PVAs can also leak additional information.
| Original language | English |
|---|---|
| Title of host publication | Computer Security. ESORICS 2021 International Workshops |
| Subtitle of host publication | |
| Editors | S. Katsikas |
| Publisher | Springer International Publishing |
| Pages | 317-327 |
| Number of pages | 12 |
| Volume | 13106 |
| DOIs | |
| Publication status | Published - 4 Oct 2021 |
| Externally published | Yes |
| Event | Computer Security: ESORICS 2021 International Workshops - Darmstadt, Germany Duration: 4 Oct 2021 → 8 Oct 2021 https://nr.no/en/publication/2000211/ |
Publication series
| Name | Lecture Notes in Computer Science |
|---|---|
| Publisher | Springer |
| Volume | LNCS, volume 13106 |
| ISSN (Print) | 0302-9743 |
| ISSN (Electronic) | 1611-3349 |
Workshop
| Workshop | Computer Security |
|---|---|
| Country/Territory | Germany |
| City | Darmstadt |
| Period | 4/10/21 → 8/10/21 |
| Internet address |
Field of Science*
- 2.2 Electrical engineering, Electronic engineering, Information engineering
Publication Type*
- 3.1. Articles or chapters in proceedings/scientific books indexed in Web of Science and/or Scopus database