paper-with-me

홈 › Papers

Leveraging ASR Pretrained Conformers for Speaker Verification through Transfer Learning and Knowledge Distillation

2023-09-06 · Danwei Cai, Ming Li

This paper explores the use of ASR-pretrained Conformers for speaker verification, leveraging their strengths in modeling speech signals. We introduce three strategies: (1) Transfer learning to initialize the speaker embedding network, improving generalization and reducing overfitting. (2) Knowledge distillation to train a more flexible speaker verification model, incorporating frame-level ASR loss as an auxiliary task. (3) A lightweight speaker adaptor for efficient feature conversion without altering the original ASR Conformer, allowing parallel ASR and speaker verification. Experiments on VoxCeleb show significant improvements: transfer learning yields a 0.48% EER, knowledge distillation results in a 0.43% EER, and the speaker adaptor approach, with just an added 4.92M parameters to a 130.94M-parameter model, achieves a 0.57% EER. Overall, our methods effectively transfer ASR capabilities to speaker verification tasks.

📄 PDF Abstract BibTeX arXiv:2309.03019

Code (0)

등록된 구현이 없습니다.

Tasks

Knowledge DistillationSpeaker VerificationTransfer Learning

Methods 이 논문이 사용한 방법론

Knowledge Distillation A very simple way to improve the performance of almost any machine learning algorithm is to train many different models on the same data and then to average their predictions.…

Similar Papers 제목 키워드 기반

Universal Pooling Method of Multi-layer Features from Pretrained Models for Speaker Verification

2024-09-12 · Jin Sob Kim, Hyun Joon Park, WooSeok Shin, Sung Won Han

Recent advancements in automatic speaker verification (ASV) studies have been achieved by leveraging large-scale pretrained networks. In this study, we analyze the approaches toward such a paradigm and underline the sign…

Speaker VerificationText-Independent Speaker Verification

Improving Neural Diarization through Speaker Attribute Attractors and Local Dependency Modeling

2025-06-05 · David Palzer, Matthew Maciejewski, Eric Fosler-Lussier

In recent years, end-to-end approaches have made notable progress in addressing the challenge of speaker diarization, which involves segmenting and identifying speakers in multi-talker recordings. One such approach, Enco…

AttributeDecoderspeaker-diarizationSpeaker Diarization

Improving Speaker Verification with Self-Pretrained Transformer Models

2023-05-17 · Junyi Peng, Oldřich Plchot, Themos Stafylakis, Ladislav Mošner 외

Recently, fine-tuning large pre-trained Transformer models using downstream datasets has received a rising interest. Despite their success, it is still challenging to disentangle the benefits of large-scale datasets and …

Speaker Verification

On Time Domain Conformer Models for Monaural Speech Separation in Noisy Reverberant Acoustic Environments

2023-10-09 · William Ravenscroft, Stefan Goetze, Thomas Hain

Speech separation remains an important topic for multi-speaker technology researchers. Convolution augmented transformers (conformers) have performed well for many speech processing tasks but have been under-researched f…

Computational EfficiencySpeech Separation

Rethinking Leveraging Pre-Trained Multi-Layer Representations for Speaker Verification

2025-12-15 · Jin Sob Kim, Hyun Joon Park, Wooseok Shin, Sung Won Han arxiv

Recent speaker verification studies have achieved notable success by leveraging layer-wise output from pre-trained Transformer models. However, few have explored the advancements in aggregating these multi-level features…

Speaker Verification