SecuVoice: A Spanish Speech Corpus for Secure Applications with Smartphones

In this paper, a new speech database, the so-called SecuVoice, is described. This database consists of utterances in Spanish of isolated digits recorded with two different smartphones: a mid-range smartphone and a high-range one. This database is intended for research on biometrics and secure applications that integrate both automatic speech recognition (ASR) and speaker recognition/verification. In this regard, both ASR and speaker verification baselines are given in this paper as reference. The experimental results show that a very high performance can be obtained on this corpus. SecuVoice will be released through ELRA (European Language Resource Association), so that speech researchers can evaluate and compare the performance of their speech-related developments and algorithms within a framework with speech signals acquired with real smartphones.