Observatorio de I+D+i UPM

Memorias de investigación
Ponencias en congresos:
Attention-based word vector prediction with LSTMs and its application to the OOV problem in ASR
Año:2019
Áreas de investigación
  • Ciencias de la computación y tecnología informática
Datos
Descripción
We propose three architectures for a word vector prediction system (WVPS) built with LSTMs that consider both past and future contexts of a word for predicting a vector in an embedded space where its surrounding area is semantically related to the considered word. We introduce an attention mechanism in one of the architectures so the system is able to assess the specific contribution of each context word to the prediction. All the architectures are trained under the same conditions and the same training material, following a curricular-learning fashion in the presentation of the data. For the inputs, we employ pretrained word embeddings. We evaluate the systems after the same number of training steps, over two different corpora composed of ground-truth speech transcriptions in Spanish language from TCSTAR and TV recordings used in the Search on Speech Challenge of IberSPEECH 2018. The results show that we are able to reach significant differences between the architectures, consistently across both corpora. The attention-based architecture achieves the best results, suggesting its adequacy for the task. Also, we illustrate the usefulness of the systems for resolving out-of-vocabulary (OOV) regions marked by an ASR system capable of detecting OOV occurrences
Internacional
Si
Nombre congreso
The 20th Annual Conference of the International Speech Communication Association - Interspeech 2019,
Tipo de participación
960
Lugar del congreso
Austria
Revisores
Si
ISBN o ISSN
1990-9772
DOI
http://dx.doi.org/10.21437/Interspeech.2019-2347
Fecha inicio congreso
15/09/2019
Fecha fin congreso
19/09/2019
Desde la página
3520
Hasta la página
3523
Título de las actas
Actas 20th Annual Conference of the International Speech Communication Association - Interspeech 2019,
Esta actividad pertenece a memorias de investigación
Participantes
  • Autor: Alejandro Coucheiro Limeres (UPM)
  • Autor: Ruben San Segundo Hernandez (UPM)
  • Autor: Javier Ferreiros Lopez (UPM)
  • Autor: Fernando Fernandez Martinez (UPM)
Grupos de investigación, Departamentos, Centros e Institutos de I+D+i relacionados
  • Creador: Grupo de Investigación: Grupo de Tecnología del Habla
  • Departamento: Ingeniería Electrónica
  • Centro o Instituto I+D+i: Centro de I+d+i en Procesado de la Información y Telecomunicaciones
S2i 2022 Observatorio de investigación @ UPM con la colaboración del Consejo Social UPM
Cofinanciación del MINECO en el marco del Programa INNCIDE 2011 (OTR-2011-0236)
Cofinanciación del MINECO en el marco del Programa INNPACTO (IPT-020000-2010-22)