CPqD Text to Speech
Customize user interaction in a natural and humanized way, using speech synthesis, converting any text to speech, in real time.
CPqD Text to Speech uses cutting-edge technology to provide high-quality voices and ensure the best user experience possible. The speech synthesis techniques we use (parametric models and unit selection) produce natural-sounding, expressive voices to be used in several different applications and environments (IVR, smartphones, computers, embedded systems, etc.).
- Applied to devices with different capacities
- Flexible integration
- Streaming support
- Multiple voices and languages in a single instance
- Allows several voices and languages in the same installation, expediting the use of the technology and simplifying interaction with different applications.
The speech synthesis technology used by CPqD Text to Speech - parametric models and unit selection - make it possible to be used with quality in computer environments with different memory capacities and processing power (smartphones, tablets, computers, IVR, etc.).
API for C/C++ and Java software development, including usage examples and documentation.
The CPqD Text to Speech API supports any application or device that can be integrated via HTTP REST, MRCP v1 and v2 or Websocket interfaces.
The generated audio can be delivered as the synthesizer produces the synthetic speech, meeting the needs of systems based on media streaming, operating in real time.
- Available voices and languages
- Rosana (female Portuguese)
- Carlos (male Portuguese)
- Paola (female Spanish)
- Use different voices in the same text The API allows different voices to be used alternately in the synthesis of a text, as well as the use of prerecorded synthetic speech audio, by means of simple SSML tags.
- Expression and sound effects
CPqD Text to Speech can be made to sound even more naturally by adding expression (different recording styles, or expressing emotions such as joy, anger or surprise, for example), by voice variations (rhythm or intonation), and sound effects (breathing, laughter, crying, etc. ). These features can be easily and speedily integrated by means of SSML tags inserted in the text itself.
Talk to one of our specialists.
CPqD develops and offers robust network analysis and structuring projects with operational and energetic efficiency, as well as predictive maintenance.Talk to a Specialist