This paper presents an analysis of the KPT system for the 2022 NIST Language Recognition Evaluation. The KPT submission focuses on the fixed training condition where only specific speech data can be used to develop all the modules and auxiliary systems used to build the language recognizer. Our solution consists of several sub-systems based on different neural network front-ends and a common back-end for classification and fusion. The goal of each front-end is to extract language-related embeddings. Gaussian linear models are used to classify the embeddings of each front-end, followed by multi-class logistic regression to calibrate and fuse the different sub-systems. Experimental results from the NIST LRE 2022 evaluation task show that our approach achieves competitive performance.

Description and analysis of the KPT system for NIST Language Recognition Evaluation 2022 / Sarni, Salvatore; Cumani, Sandro; Siniscalchi, Sabato Marco; Bottino, Andrea. - STAMPA. - (2023), pp. 1933-1937. (Intervento presentato al convegno 24th INTERSPEECH Conference 2023 tenutosi a Dublin (IRL) nel 20th – 24th August 2023) [10.21437/Interspeech.2023-155].

Description and analysis of the KPT system for NIST Language Recognition Evaluation 2022

Sarni, Salvatore;Cumani, Sandro;Bottino, Andrea
2023

Abstract

This paper presents an analysis of the KPT system for the 2022 NIST Language Recognition Evaluation. The KPT submission focuses on the fixed training condition where only specific speech data can be used to develop all the modules and auxiliary systems used to build the language recognizer. Our solution consists of several sub-systems based on different neural network front-ends and a common back-end for classification and fusion. The goal of each front-end is to extract language-related embeddings. Gaussian linear models are used to classify the embeddings of each front-end, followed by multi-class logistic regression to calibrate and fuse the different sub-systems. Experimental results from the NIST LRE 2022 evaluation task show that our approach achieves competitive performance.
File in questo prodotto:
File Dimensione Formato  
sarni23_interspeech.pdf

accesso aperto

Tipologia: 2a Post-print versione editoriale / Version of Record
Licenza: Pubblico - Tutti i diritti riservati
Dimensione 226.42 kB
Formato Adobe PDF
226.42 kB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2979479