The Toshiba Mandarin TTS System for the Blizzard Challenge 2009
نویسندگان
چکیده
This paper introduces the Toshiba Mandarin Text-to-Speech (TTS) system submitted to the Mandarin benchmark of the Blizzard Challenge 2009. The basic framework keeps unchanged with the system in 2008 and we modify the system in several aspects: automatically find bad units in the database when preparing the speech corpus, add a G2P procedure after the text analysis to increase accuracy of the predicted pinyin for heteronyms, introduce prosody layer information into the prosody modeling and modify the fusion methods to fuse units in frequency domain. The subjective evaluation results show that these modifications improve the performance.
منابع مشابه
The Toshiba Mandarin TTS System for the Blizzard Challenge 2008
This paper describes the Toshiba Mandarin Text-to-Speech (TTS) system that was submitted to the Blizzard Challenge 2008. The front-end of the system uses machine-learning approaches such as generalized linear models (GLM) and Quantification Method Type 1 (QMT1) to predict pause, duration and F0 contour. According to the predicted prosody information, the back-end of the system uses Toshiba’s ow...
متن کاملMultilingual MARY TTS participation in the Blizzard Challenge 2009
The paper describes the Blizzard Challenge 2009 participation of MARY TTS, an open-source TTS system using a unit selection voice. We briefly outline the new language support framework we provide so that people can add support for their languages to MARY TTS, and describe how that framework was used for building a Mandarin Chinese system and voice. The system performs well for English and reaso...
متن کاملThe UPC TTS System Description for the 2008 Blizzard Challenge
This paper presents the UPC TTS system named Ogmios. It was used to generate the voices in UK English and Mandarin Chinese for Blizzard Challenge 2008. Ogmios is a system based on unit-selection using acoustic and phonetic features both in target and concatenation costs. Most of the modules of Ogmios rely on data driven techniques. This evaluation confirms that this framework allows fast develo...
متن کاملThe NTUT Blizzard Challenge 2009 Entry
This paper describes the process of building HMM-based speech synthesis system (HTS) voices for our participation in the Blizzard Challenge 2009. Out of the two languages required (English and Mandarin Chinese) we only built three Mandarin Chinese voices for main hub (MH) and two spoke (MS1 and MS2) tasks. According to the evaluation results, our MH voice got 3 points for both mean opinion scor...
متن کاملThe WISTON Text - to - Speech System for Blizzard Challenge 2009
This paper describes the WISTON system, a large corpus based TTS system that was submitted to Blizzard Challenge 2009. The text analysis part of this system contains text preprocessing, word segmentation, POS tagging, phonetic transcription and prosody structure prediction, most of which are based on Maximum Entropy (ME) models. In unit selection part, CART models are used to predict the prosod...
متن کامل