A Spoken Document Retrieval System for TV Broadcast News in Spanish and Basque

• Autores: Amparo Varona Fernández , Silvia Nieto, Luis Javier Rodríguez Fuentes, Mikel Peñagaricano Badiola, Germán Bordel García , Mireia Díez Sánchez
• Localización: Procesamiento del lenguaje natural, ISSN 1135-5948, Nº. 47, 2011, págs. 75-83
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen
  • español

    El sistema de indexado y búsqueda de contenidos multimedia que se presenta en este trabajo (Hearch) es un buscador de aspecto convencional pero con la capacidad de devolver segmentos de vídeo gracias a la transcripción automática de sus contenidos de voz. El sistema consta de un back-end que capta, procesa e indexa los recursos, y de un front-end que permite realizar búsquedas y configurar y monitorizar el funcionamiento de los distintos módulos, mediante una interfaz web. Actualmente se encuentra operativa una versión de la herramienta que trabaja frente a repositorios de noticias en castellano y euskera (http://gtts.ehu.es/Hearch/). Para evaluar el rendimiento del sistema se dispone de 6 programas de noticias en castellano y 7 en euskera. Puesto que el módulo de Reconocimiento Automático del Habla introduce bastantes errores, se ha propuesto y evaluado una aproximación basada en añadir términos afines a los de la pregunta para ampliar los resultados proporcionados por el sistema. Como resultado se obtiene una pequeña mejora del rendimiento.

  • English

    This paper presents a spoken document retrieval system (Hearch) looking like a conventional search tool, which retrieves audio/video segments based on the automatic transcription of speech contents. The system consists of a back-end that captures, processes and indexes audio/video resources, and a front-end that allows to search contents, configure various modules and display performance statistics through a web interface. An early version of this tool is available (http://gtts.ehu.es/Hearch/), which searches and retrieves segments on TV broadcast news repositories in Spanish and Basque. To evaluate the performance of the system, six manually transcribed TV broadcast news in Spanish and seven in Basque have been used. An approach based on extending the query with the so called friendly terms has been proposed and evaluated, attempting to minimize the effect of errors introduced by the Automatic Speech Recognition module. This approach led to slight performance improvements.

• Referencias bibliográficas
  • Aduriz, I., E. Agirre, I. Aldezabal, I. Alegria, O. Ansa, X. Arregi, J.M. Arriola, X. Artola, A. Diaz de Ilarraza, N. Ezeiza, K. Gojenola,...
  • Alberti, C., M. Bacchiani, A. Bezman, C. Chelba, A. Drofa, H. Liao, P. Moreno, T. Power, A. Sahuguet, M. Shugrina, and O. Siohan. 2009. An...
  • Atserias, Jordi, Bernardino Casas, Elisabet Comelles, Meritxell González, Lluís Padró, and Muntsa Padró. 2006. Free- Ling 1.3: Syntactic and...
  • Bordel, G., A. Casillas, M. Penagarikano, L.J. Rodriguez-Fuentes, and A. Varona. 2009. An XML Resource Definition for Spoken Document Retrieval....
  • Clements, M. and M. Gavalda. 2007. Voice/audio information retrieval: minimizing the need for human ears. In Proc. of IEEE ASRU Workshop,...
  • Diez, M., M. Penagarikano, A. Varona, L.J. Rodriguez-Fuentes, and G. Bordel. 2011. On the use of dot scoring for speaker diarization. In Iberian...
  • Frakes, W.B. and R. Baeza-Yates. 1992. Information Retrieval. Prentice Hall.
  • Glass, James R., Timothy J. Hazen, D. Scott Cyphers, Igor Malioutov, David Huynh, and Regina Barzilay. 2007. Recent progress in the MIT spoken...
  • Hansen, J. H. L. et al. 2005. SpeechFind: Advances in Spoken Document Retrieval for a National Gallery of the Spoken Word. IEEE Transactions...
  • Hatcher, Erik, Otis Gospodnetic, and Mc-Candless M. 2010. Lucene in Action. Manning Publications Co. 2nd edition.
  • Jelinek, Frederick. 1999. Statistical Methods for Speech Recognition (Second Edition). Language, Speech and Communication Series. The MIT...
  • Kiranyaz, S., Ahmad Farooq Qureshi, and M. Gabbouj. 2006. A generic audio classification and segmentation approach for multimedia indexing...
  • Lee, Donghyeon and Gary Geunbae Lee. 2008. A Korean Spoken Language Document Retrieval System for Lecture Search. In SCSS.
  • Makhoul, J., F. Kubala, T. Leek, D. Liu, L. Nguyen, R. Schwartz, and A. Srivastava. 2000. Speech and Language Technologies for Audio Indexing...
  • Mamou, Jonathan and Bhuvana Ramabhadran. 2008. Phonetic query expansion for spoken document retrieval. In Proc. Interspeech.
  • Mills, Timothy J., David Pye, Nicholas J. Hollinghurst, and Kenneth R. Wood. 2000. AT&TV: Broadcast Television and Radio Retrieval. In...
  • Moreno, Asuncion, Dolors Poch, Antonio Bonafonte, Eduardo Lleida, Joaquim Llisterri, Jose B. Marino, and Climent Nadeu. 1993. Albayzin speech...
  • Ohtsuki, Katsutoshi, Katsuji Bessho, Yoshihiro Matsuo, Shoichi Matsunaga, and Yoshihiko Hayashi. 2006. Automatic Multimedia Indexing. IEEE...
  • Penagarikano, M. and G. Bordel. 2005. Sautrela: A Highly Modular Open Source Speech Recognition Framework. In Proceedings of the IEEE ASRU...
  • Rodriguez-Fuentes, L.J., M. Penagarikano, A. Varona, M. Diez, and G. Bordel. 2010. GTTS Systems for the Albayzin 2010 Audio Segmentation Evaluation....
  • Siemund, R., H. Höge, S. Kunzmann, and K. Marasek. 2000. SPEECON - speech data for consumer devices. In Proc. LREC, pages 883–886.
  • Stolcke, Andreas. 2002. SRILM - an extensible language modeling toolkit. In Proceedings of ICSLP, pages 257–286.
  • Thong, J.M. Van, P.J. Moreno, B. Logan, B. Fidler, K. Maffey, and M. Moores. 2002. SpeechBot: An Experimental Speech-Based Search Engine for...
  • Varona, A., Penagarikano M., Rodriguez-Fuentes L.J., M. Diez, and G. Bordel. 2010. Verification of the four Spanish official languages on...
  • Ye, Ruizhi, Yingchun Yang, Zhenyu Shan, Yiyan Liu, and Sen Zhou. 2006. ASEKS: A P2P Audio Search Engine Based on Keyword Spotting. In Proceedings...
  • Young, S. et al. 2006. The HTK Book (Version 3.4). Cambridge, UK.