PromptSculptor™ is a powerful and very flexible text-to-speech tuning tool for RealSpeak. With its intuitive Graphical User Interface (GUI) it allows a developer to quickly create natural sounding prompts tailored to a specific application context and nearly undistinguishable from recorded prompts. PromptSculptor does not require a deep linguistic knowledge or TTS expertise and significantly reduces application development time and voice talent costs.

Tuned results can be stored in ActivePrompts™ databases at only a fraction of the storage cost of wave files. Nuance's ActivePrompts technology allows to re-use tuning results for entire sentences, phrases and sub-phrases in the run-time RealSpeak Solo engine.

ActivePrompts also provides a unique mechanism to seamlessly blend tuned carrier phrases (e.g. "Turn left on to..." ) with variable content slots (e.g. "Main Street" or "Fifth Avenue") in the run-time engine.

new PromptSculptor now also offers Speechbase Customization. Even with the smallest embedded speechbase, high quality prompts can be created by selective addition of speech segments. Maximum quality tuned prompts can thus be obtained with minimal footprint increase.

• Intuitive Graphical User Interface (GUI).
• Alternative pronunciations can be generated on-demand.
• Supports phonetic editing and unit selection.
• new Tuning can be done on an extended database providing high-quality prompts comparable to recorded prompts, with minimal adjustments.
• new The standard deployed database is expanded with the new units required for the prompts.
• new All tuned prompts are stored as ActivePrompts™ and used systematically a runtime.
• Compatible with all RealSpeak Solo voices.
• Support for SSML.
• Output speech files can be generated at 8, 11, 16, 20 kHz.
• Waveform format supported include riff, raw, a-law, u-law.

• RealSpeak Solo allows common phrases (carrier phrases) to be generated with high quality approaching recorded prompts.
• Seamless combination of high quality phrases and dynamically generated content, ideal for applications such as navigation or speech-controlled music playing.
• Text-to-speech can be used in more applications where high quality output is required.
• Tuning of TTS output is reduced significantly thus shortening time to market.

PromptSculptor adds expressiveness to your prompts

Before:	Sound sample		(480 KB)
After:	Sound sample		(480 KB)

PromptSculptor improves the quality of carrier phrases

Before:	Sound sample		(480 KB)
After:	Sound sample		(480 KB)