DNN and i-vector combined method for speaker recognition on multi-variability environments
| Autor: | Flavio J. Reyes-Díaz, Gabriel Hernández-Sierra, José Ramón Calvo de Lara |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Linguistics and Language
Reverberation Artificial neural network Computer science Speech recognition Speaker recognition Language and Linguistics Compensation (engineering) Human-Computer Interaction Discriminative model Robustness (computer science) Computer Vision and Pattern Recognition Representation (mathematics) Environmental noise Software |
| Zdroj: | International Journal of Speech Technology. 24:409-418 |
| ISSN: | 1572-8110 1381-2416 |
| DOI: | 10.1007/s10772-021-09796-1 |
| Popis: | The article deals with the compensation of variability in Automatic Speaker Verification systems in scenarios where the variability conditions due to utterance duration, reverberation and environmental noise are simultaneously present. We introduce a new representation of the speaker’s discriminative information, based on the use of a deep neural network trained discriminatively for speaker classification and i-vector representation. The proposed representation allows us to increase the verification performance by reducing the error between 2.5 and 7.9 % for all variability conditions compared to baseline systems. We also analyze the speaker verification system robustness based on interquartile range, obtaining a 1.19 times improvement compared to baselines evaluated. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink