On-device Speaker Anonymization Of Acoustic Embeddings For ASR Based Onflexible Location Gradient Reversal Layer

Abstract

Smart devices serviced by large-scale AI models necessitates user data transfer to the cloud for inference. For speech applications, this means transferring private user information, e.g., speaker identity. Our paper proposes a privacy-enhancing framework that targets speaker identity anonymization while preserving speech recognition accuracy for our downstream task~-~Automatic Speech Recognition (ASR). The proposed framework attaches flexible gradient reversal based speaker adversarial layers to target layers within an ASR model, where speaker adversarial training anonymizes acoustic embeddings generated by the targeted layers to remove speaker identity. We propose on-device deployment by execution of initial layers of the ASR model, and transmitting anonymized embeddings to the cloud, where the rest of the model is executed while preserving privacy. Experimental results show that our method efficiently reduces speaker recognition relative accuracy by 33%, and improves ASR performance

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyjalal2023on

Related papers

• Reprogramming Self-supervised Learning-based Speech Representations For Speaker Anonymization (2023)2.26
• Asynchronous Voice Anonymization Using Adversarial Perturbation On Speaker Embedding (2024)7.16
• Language-independent Speaker Anonymization Approach Using Self-supervised Pre-trained Models (2022)9.92
• Privacy-utility Balanced Voice De-identification Using Adversarial Examples (2022)0.00
• A Method To Reveal Speaker Identity In Distributed ASR Training, And How To Counter It (2021)5.84
• A Speech Representation Anonymization Framework Via Selective Noise Perturbation (2022)6.34
• Speaker De-identification System Using Autoencoders And Adversarial Training (2020)0.00
• Privacy Attacks For Automatic Speech Recognition Acoustic Models In A Federated Learning Framework (2021)9.23