A framework for speaker retrieval and identification through unsupervised learning

Autor(es)

Campos, Victor de Abreu ^[1] ; Guimaraes Pedronette, Daniel Carlos ^[1]

Afiliação do(s) autor(es)

^[1] State Univ Sao Paulo, UNESP, Dept Stat Appl Math & Comp, Rio Claro - Brazil

• Tipo de documento: Artigo Científico
• Fonte: COMPUTER SPEECH AND LANGUAGE; v. 58, p. 153-174, NOV 2019.
• Citações Web of Science: 0

Resumo

Speaker recognition is a task of remarkable relevance, with applications in diversified domains. Recently, mainly due to the facilities in audio-visual content acquisition, the capacity of analyzing growing datasets independent of labeled data has become a crucial advantage. This paper presents a speaker recognition approach based on recent unsupervised learning methods, which do not require any labeled data or user intervention. The approach is organized in terms of a framework which exploits a rank-based formulation. The similarity information defined by speaker modeling techniques is encoded in ranked lists, which are used as input by the unsupervised learning algorithms. Vector quantization, Gaussian mixture models and i-vectors are employed as modeling techniques, while the algorithms RL-Sim and ReckNN are used for unsupervised learning tasks. The framework was experimentally evaluated on query-by-example speaker retrieval and speaker identification tasks, both on clean and noisy speech recordings. An experimental evaluation was conducted on three public datasets, different languages, and recordings conditions. Effectiveness gains up to +56% on retrieval measures were obtained through the use of unsupervised learning algorithms over traditional speaker recognition techniques. (C) 2019 Elsevier Ltd. All rights reserved. (AU)

• Processo FAPESP: 17/25908-6 - Aprendizado fracamente supervisionado para análise de vídeos no domínio comprimido em tarefas de recuperação e classificação para alertas visuais
• Beneficiário: João Paulo Papa
• Modalidade de apoio: Auxílio à Pesquisa - Parceria para Inovação Tecnológica - PITE
• Processo FAPESP: 15/07934-4 - Identificação automática de locutor utilizando métodos de aprendizado não-supervisionado
• Beneficiário: Victor de Abreu Campos
• Modalidade de apoio: Bolsas no Brasil - Mestrado
• Processo FAPESP: 18/15597-6 - Aplicação e investigação de métodos de aprendizado não-supervisionado em tarefas de recuperação e classificação
• Beneficiário: Daniel Carlos Guimarães Pedronette
• Modalidade de apoio: Auxílio à Pesquisa - Jovens Pesquisadores - Fase 2