VoicePAT: An Efficient Open-Source Evaluation Toolkit for Voice Privacy Research

Abstract

Speaker anonymization is the task of modifying a speech recording such that the original speaker cannot be identified anymore. Since the first Voice Privacy Challenge in 2020, along with the release of a framework, the popularity of this research topic is continually increasing. However, the comparison and combination of different anonymization approaches remains challenging due to the complexity of evaluation and the absence of user-friendly research frameworks. We therefore propose an efficient speaker anonymization and evaluation framework based on a modular and easily extendable structure, almost fully in Python. The framework facilitates the orchestration of several anonymization approaches in parallel and allows for interfacing between different techniques. Furthermore, we propose modifications to common evaluation methods which improves the quality of the evaluation and reduces their computation time by 65 to 95%, depending on the metric. Our code is fully open source.

Speaker Anonymization: Disentangling Speaker Features from Pre-Trained Speech Embeddings for Voice Conversion

“…The experimental results validate the approach, although a more comprehensive analysis on the quality of the generated signals is suggested. Similarly, a specific study on residual speaker characteristics embedded in the learned representation would permit a fine-grained control of the converted speech when compared to the very recent VoicePAT framework [81] selected for the incoming VoicePrivacy Challenge 2024.…”