Enhancing Speaker Verification through Trainable Audio Features

Xylogiannis P, Emvoliadis A, Vryzas N, Vrysis L, Dimoulas C. Enhancing Speaker Verification through Trainable Audio Features. In: Express Paper. Audio Engineering Society; 2024. Paper 245. Available from: https://aes.org/publications/elibrary-page/?id=22591

@inproceedings{Xylogiannis2024_22591,
  author = {Xylogiannis, Paris and Emvoliadis, Alexandros and Vryzas, Nikolaos and Vrysis, Lazaros and Dimoulas, Charalampos},
  title = {{Enhancing Speaker Verification through Trainable Audio Features}},
  note = {Paper 245},
  year = {2024},
  month = jun,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=22591}
}

TY  - CPAPER
TI  - Enhancing Speaker Verification through Trainable Audio Features
AU  - Xylogiannis, Paris
AU  - Emvoliadis, Alexandros
AU  - Vryzas, Nikolaos
AU  - Vrysis, Lazaros
AU  - Dimoulas, Charalampos
M1  - Paper 245
PY  - 2024
DA  - 2024/06/06
UR  - https://aes.org/publications/elibrary-page/?id=22591
PB  - Audio Engineering Society
LA  - en
AB  - This study presents a novel pipeline for speaker verification, exploiting the advantages of trainable audio features to enhance identification accuracy and efficiency. Unlike traditional systems that rely heavily on spectrograms and handcrafted features like Mel-Frequency Cepstral Coefficients (MFCCs), our approach incorporates a learnable audio frontend encoder, a speaker embeddings system, and a decoder for signal reconstruction. This pipeline employs audio encoding methods to directly process raw audio inputs, dynamically adapting its filters to extract reliable speaker-specific features. By incorporating trainable features, our system not only captures a broader range of speaker information but also offers flexibility in handling either direct output from the audio encoder or signals reconstructed by the decoder. This capability allows for an in-depth analysis of the impact of direct feature extraction on speaker identification tasks. The proposed pipeline is evaluated on a frequently used speaker verification dataset. Our findings demonstrate significant improvements over the baselines feature extraction method, highlighting the potential of trainable features in redefining the landscape of speech processing systems.
ER  -