Core Tech

Features of jTTS v3.0

jTTS v3.0 is the core of the main products sold by SinoVoice. As primary speech products on market, this core is of following features:

•  Supporting mixed reading of Chinese and English. The English words are of the same timbre as the Chinese words.

•  Supporting various character sets such as GB2312, GBK, Big5 and Unicode. In particular, it can read all the 20,000 characters of GBK including traditional and rare Chinese characters.

•  High preprocessing capability of text analysis. Figures, English words and polyphonic characters can be pronounced more accurate.

•  Unique analysis technology of hierarchical prosody structure and hierarchical intonation and rhythm model, which makes the synthesized speech more natural and fluent.

•  Supporting the setting of various synthesis parameters.

•  Supporting text tag which can generate better synthesis effect.

•  Providing dictionary modification tool by which users can add phrases and phonetic notations in specific fields.

•  Supporting cocurrent TTS of multi channels. Along with performance optimization, it can satisfy the requirements for large-scale speech synthesis.

Features of jTTS v4.0 <<