Evaluating Novel Speech Transcription Architectures on the Spanish RTVE2020 Database

Applied Sciences

Ander González-Docasal¹  Aitor Álvarez¹  Iván G. Torre¹  Haritz Arzelus¹ 

[1] Vicomtech Foundation, Basque Research and Technology Alliance (BRTA), Mikeletegi 57, 20009 Donostia-San Sebastián, Spain;

关键词: automatic speech recognition;    deep learning;    Spanish;    convolutional neural networks;    recurrent neural networks;    embedded systems;   

DOI  :  10.3390/app12041889

来源: DOAJ

【 摘 要 】

This work presents three novel speech recognition architectures evaluated on the Spanish RTVE2020 dataset, employed as the main evaluation set in the Albayzín S2T Transcription Challenge 2020. The main objective was to improve the performance of the systems previously submitted by the authors to the challenge, in which the primary system scored the second position. The novel systems are based on both DNN-HMM and E2E acoustic models, for which fully- and self-supervised learning methods were included. As a result, the new speech recognition engines clearly outperformed the performance of the initial systems from the previous best WER of 19.27 to the new best of 17.60 achieved by the DNN-HMM based system. This work therefore describes an interesting benchmark of the latest acoustic models over a highly challenging dataset, and identifies the most optimal ones depending on the expected quality, the available resources and the required latency.

【 授权许可】

Unknown