Progress in the BBN 2007 Mandarin Speech to Text system

Autor:	Tim Ng, Bing Zhang, Kham Nguyen, Long Nguyen
Rok vydání:	2008
Předmět:	Computer science business.industry Speech recognition Word error rate computer.software_genre Mandarin Chinese language.human_language Reduction (complexity) language Artificial intelligence Pitch tracking business computer Natural language Natural language processing
Zdroj:	ICASSP
ISSN:	1520-6149
DOI:	10.1109/icassp.2008.4517915
Popis:	In this paper, we describe the BBN 2007 Mandarin speech-to-text system developed for the GALE Evaluation 2007. In comparison to the BBN 2006 Mandarin system, we achieved 25% relative reduction in character error rate on the most important test sets. The utilization of all available training data provided the largest contribution to the improvement. The use of a better pitch tracking algorithm also contributed significantly, while system combination made some noticeable improvement too.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::d6584b7072ef79cf5f568b6f42378c98 https://doi.org/10.1109/icassp.2008.4517915 Zobrazit plný text záznamu