Phonetic-Oriented Identification of Twin Speakers Using 4-Second Vowel Sounds and a Combination of a Shift-Invariant Phase Feature (NRD), MFCCs, and F0 Information

Ferreira A. Phonetic-Oriented Identification of Twin Speakers Using 4-Second Vowel Sounds and a Combination of a Shift-Invariant Phase Feature (NRD), MFCCs, and F0 Information. In: AES Conference: 2019 AES International Conference on Audio Forensics. Audio Engineering Society; 2019.pp. 13:1-17. Available from: https://aes.org/publications/elibrary-page/?id=20466

@inproceedings{Ferreira2019_20466,
  author = {Ferreira, Anìbal},
  title = {{Phonetic-Oriented Identification of Twin Speakers Using 4-Second Vowel Sounds and a Combination of a Shift-Invariant Phase Feature (NRD), MFCCs, and F0 Information}},
  booktitle = {AES Conference: 2019 AES International Conference on Audio Forensics},
  pages = {13:1-17},
  url = {https://aes.org/publications/elibrary-page/?id=20466},
  year = {2019},
  month = jun,
  publisher = {Audio Engineering Society}
}

TY  - CPAPER
TI  - Phonetic-Oriented Identification of Twin Speakers Using 4-Second Vowel Sounds and a Combination of a Shift-Invariant Phase Feature (NRD), MFCCs, and F0 Information
AU  - Ferreira, Anìbal
T2  - AES Conference: 2019 AES International Conference on Audio Forensics
SP  - 1
EP  - 17
M1  - Paper 13
PY  - 2019
DA  - 2019/06/06
UR  - https://aes.org/publications/elibrary-page/?id=20466
PB  - Audio Engineering Society
LA  - en
AB  - Automatic speaker identi?cation typically relies on sophisticated statistical modeling and classi?cation which requires large amounts of data for good performance. However, in actual audio forensics casework, frequently only a few seconds of speech material are available. In this paper, we favor diversity in feature extraction, simple modeling and classi?cation, and constructive combination of congruent classi?cation scores. We use phase, spectral magnitude and F0-related features in speaker identi?cation experiments on a database of 35 speakers most of whom are twins. Using only 4.4 sec. of vowel-like sounds per speaker, we characterize the performance that is reached with individual features and we characterize simple and yet effective ways of classi?cation score fusion. Insights for further research are also presented.
ER  -