Efficient Black-box Speaker Verification Model Adaptation With Reprogramming And Backend Learning

Abstract

The development of deep neural networks (DNN) has significantly enhanced the performance of speaker verification (SV) systems in recent years. However, a critical issue that persists when applying DNN-based SV systems in practical applications is domain mismatch. To mitigate the performance degradation caused by the mismatch, domain adaptation becomes necessary. This paper introduces an approach to adapt DNN-based SV models by manipulating the learnable model inputs, inspired by the concept of adversarial reprogramming. The pre-trained SV model remains fixed and functions solely in the forward process, resembling a black-box model. A lightweight network is utilized to estimate the gradients for the learnable parameters at the input, which bypasses the gradient backpropagation through the black-box model. The reprogrammed output is processed by a two-layer backend learning module as the final adapted speaker embedding. The number of parameters involved in the gradient calculation is sma

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyli2023efficient

Related papers

• An Investigation Of Reprogramming For Cross-language Adaptation In Speaker Verification Systems (2024)2.26
• Speaker Verification Using End-to-end Adversarial Language Adaptation (2018)11.19
• Vae-based Domain Adaptation For Speaker Verification (2019)7.50
• Adapting End-to-end Neural Speaker Verification To New Languages And Recording Conditions With Adversarial Training (2018)9.59
• DEAAN: Disentangled Embedding And Adversarial Adaptation Network For Robust Speaker Representation Learning (2020)9.59
• SE/BN Adapter: Parametric Efficient Domain Adaptation For Speaker Recognition (2024)0.00
• Self-supervised Learning Based Domain Adaptation For Robust Speaker Verification (2021)11.49
• Bayesian Learning For Deep Neural Network Adaptation (2020)9.76