Abstract
This paper investigates the auditory and acoustical characteristics of whispery disguised voice, and compares the patterns with those of normal (non-disguised) voices. It also evaluates effects of whispery disguise on forensic voice comparison. Recordings of eleven male college students' normal voices and whispery disguised voices were collected. All their normal and whisper speech was acoustically analyzed and compared. The parameters including average syllable duration, intensity, vowel formant frequencies, and long term average spectrum (LTAS) were measured and statistically analyzed. The effect of whispery voice disguise on speaker recognition by auditory perception and an automatic system were evaluated. Correlation and regression analyses were made on the parameters of whispery voice and normal voice. These simple regression models can be used for parameter compensation in forensic casework.
| Original language | English |
|---|---|
| Pages (from-to) | 1413-1416 |
| Number of pages | 4 |
| Journal | Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH |
| Volume | 2018-September |
| DOIs | |
| Publication status | Published - Sept 2018 |
| Event | 19th Annual Conference of the International Speech Communication, INTERSPEECH 2018 - Hyderabad, India Duration: 2 Sept 2018 → 6 Sept 2018 |
Keywords
- Chinese
- Forensic speaker recognition
- Voice disguise
- Whisper
ASJC Scopus subject areas
- Language and Linguistics
- Human-Computer Interaction
- Signal Processing
- Software
- Modelling and Simulation