Stunningly accurate, embedded and cloud-based voice recognition.
Cerence’s new generation of text-to-speech solutions transforms the voice assistant experience by offering the most natural text-to-speech for every cloud and embedded use case.
Cerence TTS is a suite of speech output solutions to generate high-quality speech, with seamless blending of dynamic text-to-speech, pre-recorded audio, and tuned prompts. Cerence TTS is optimized to read long texts in a natural, human way.
New deep learning-based algorithms deliver higher smoothness and more natural prosody, resulting in a unique voice experience. Cerence TTS is available through distribution partners for automotive and non- automotive applications such as transportation, accessibility, consumer electronics, industrial logistics and public announcements.
Features and Benefits
Multiple Languages and Voices
A truly universal voice portfolio offers 64 languages and 141 voices for the creation of global solutions using a single engine. The language and voice portfolio are continually expanding
Developers can choose from 4 different speaking styles: neutral, lively, forceful, and empathic
Gilded Speech Databases
Speaking styles are enhanced by selecting expressive pre- recorded prompts (incl. non- verbal) from a “gilded speech” database accompanying the TTS voice
Gender Neutral Voices
A new timbre markup tag controls the perceived age, gender, or physical size of a TTS voice
Automatic language identification, foreign language dictionaries, and high-quality acoustic extensions provide unparalleled multi-lingual readout
Use markup tags to improve the readout of an address, date, phone number, number, spelling, currency, and more
Volume, pitch, speaking rate, and timbre can be changed at run time for more dynamic and lively effects
Optimize quality using phonetic information from an external contents source like music or map data
Create your own dictionaries for out-of- vocabulary words
Recorded audio prompts or tuned prompts are seamlessly blended with dynamic text-to-speech using automatic text matching (active prompt mechanism)
Cerence TTS Designer
A comprehensive Windows tool to create TTS tuning resources, like user text rules, user dictionaries, and prompt databases
High linguistic accuracy offers correct readout for all types of text including a large set of personal names
A wide range of embedded footprints scaling from 8 to over 900 MB ensures optimal performance from low-cost platforms to powerful multimedia systems
Daily dictionary updates are offered through Cerence Cloud, guaranteeing accurate pronunciation of trending words/names in the news domain.