| ||||||||
| Text-to-speech (TTS) engines have become more and more natural in the past years. Just a few years ago, their purpose was to deliver the content of a text. This goal has been achieved: recent TTS engines can read almost any text in a comprehensive and natural sounding way. Nowadays, customers require highly sophisticated TTS engines with a natural voice, reflecting the prosody of a human utterance and pronouncing every word correctly. The innovative ELeX text-to-speech engines focus on these aspects. ELeX covers a variety of different applications for TTS: For technical specifications see the detailled product descriptions. For further information, please contact us.Multi MediaFor multi media applications, the TTS engines incorporate a high sample rate in order to match excellent sound quality. 16 kHz or 22 kHz are the highest sample rates available. TelephonyTelephony servers, however, must be able to respond to telephone calls in realtime and serve many telephone channels simultaneously. The sample output does not necessarily need to support high frequencies, since a telephone line only supports 8 kHz. ELeX offers these products for the telephony market:EmbeddedOn embedded systems, TTS can be extremely useful for various purposes like navigation, news reading, and many others, where the small display of a handheld is inappropriate. TTS on embedded systems requires a much smaller amount of resources - memory, CPU time and hard disk space usage must be as low as possible.
ToolsELeX offers additional tools for easy integration of TTS. A batch conversion tool is TextSpeak. We keep your lexicons up-to-date with LexServe: your synthesized texts are checked perpetually for new words. | |||||||