Robust Speaker Recognition for Whispered Speech
Authors:
- 26 March 2025
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
Abstract
Although significant progress has been made in developing stable and robust speaker verification systems, they still have some limitations. When a speaker's style changes, it can affect the performance of the verification system. In this paper, we consider the whispered speech domain. In particular, the adaptation of the verification system based on wav2vec 2.0 is proposed in order to achieve high quality in both the whispered and normal speech domains. Augmentation techniques based on whisper synthesis using Praat software, along with a domain score normalization approach, are utilized to improve performance in the whispered domain. Additionally, a method for whisper detection is employed. The proposed approaches significantly enhance system performance on whispered data.Keywords
This publication has 22 references indexed in Scilit:
- SR-HuBERT : An Efficient Pre-Trained Model for Speaker Verification2024
- Voxblink: A Large Scale Speaker Verification Dataset on Camera2024
- The effects of whispered speech on state-of-the-art voice based biometrics systems2015
- Whispered speaker verification and gender detection using weighted instantaneous frequencies2013
- Audiovisual corpus to analyze whisper speech2013
- Speaker identification with whispered speech using unvoiced-consonant phonemes2012
- Measurement of Speech Effort During Fluency-Inducing Conditions in Adults Who Do and Do Not StutterJournal of Speech, Language, and Hearing Research, 2009
- Acoustic Analysis of Consonants in Whispered SpeechJournal of Voice, 2008
- Whispering Speaker Identification2007
- Analysis and recognition of whispered speechSpeech Communication, 2005