Cepstral Launches 6.0 Release of Voices for Open Source Telephony
Cepstral has released new text-to-speech voices for telephony applications, created using its upgraded text-to-speech engine, Cepstral Version 6.0.
The new voice package includes barge-in functionality as well as clearer sound processing optimized for telephony equipment use. Users will notice increased voice clarity in telephony installations as well as significant improvements in breadth of lexicon coverage.
Cepstral 6.0 was designed with Asterisk users in mind.
New features include:
- Dramatically improved audio quality. The audio in Cepstral 6.0 has been tuned for crystal clarity over a phone.
- Optimized unit selection. Cepstral 6.0 represents a breakthrough in predicting and protecting speech context. This results in more natural-sounding speech with far fewer errors.
- Ultrasmooth joins and natural prosody: Cepstral 6.0 introduces a new algorithm for joining units that provides smooth and natural prosody.
Asterisk integration audio has been level matched to Asterisk prerecorded prompts for seamless integration into an Asterisk IVR/PBX. The licensing system for concurrent synthesis has been completely revamped, and is more reliable and stable.
"There are over one million Asterisk telephony implementations, and use of telephone automation and text-to-speech information is only increasing," said Dan Bagley, Cepstral CEO, in a statement. "This upgrade is very significant for all of our telephony customers and partners."