Acapela Group releases new Acapela TTS version for developers: update for new voices and Audio Boost capability.

Stay tuned and update your Acapela TTS SDK!
New voices, Audio boost, enhanced spelling mode and linguistic improvements are part of the menu of this new version. Available for Windows, Mac OSX, iOS, Android, Linux Embedded, Win CE, Windows and Linux Server.

New voices are joining the Acapela family

French, Norwegian and Swedish children voices have been added to V. 9.4. The full family of girl and boy voices is now available in our products with a unique portfolio of 18 authentic voices that keep giving the say to young users around the world.

 

• French: Elise & Valentin

• Norwegian: Emilie & Elias

• Swedish: Freja & Filip

• Anais, a new French voice has join the portfolio bringing to 9 the number of voices for this language. She adds a commercial telephone operator tone to our voice portfolio.

• And ‘Claudia Smile’ adds a cheerful voice to our German offer.  Our German voice talent has been recorded a second time under the watchfulness of our linguist experts to create a smiling version of the already existing ‘Claudia’ standard voice. Claudia will now provide interfaces and contents a pretty smile.

Speak volume with audio Boost

Audio boost provides +3dB extra loudness, for further volume expansion when it is activated to the maximum level, you can activate this feature for more volume, and this is how it works with the audio boost tag: “\audioboost=val\” where val is a value in the range 0 to 90. Try it in our online demo!

 

All Acapela SDKs are benefiting from this new version.

• In version 9.400 of Acapela TTS for Windows, Mac OSX
• In version 1.800 of Acapela TTS for iOS
• In version 1.600 of Acapela TTS for Android
• In version 8.500 of Acapela TTS for Linux Embedded
• In version 8.500 of Acapela TTS for Win CE
• In version 8.200 of Acapela TTS for Windows and Linux Server

 

> Belgian version of each French voice: including another preprocessing lexicon in order to pronounce 7x and 9x variants. On the long run it will allow alternative pronunciations as well.

 

> Enhancement of spelling mode, e.g.: \rms=1\ α => its says the full word alpha instead of spelling it “A, L, P, H, A”

 

> Memory usage optimization

 

> Enhanced synchronization on words especially for groups of words being highlighted as a single word that are now highlighted word by word.

General Linguistic Improvements

• Norwegian: Major update. This new release provides an updated Part-Of-Speech tagger as well as new breathing model and enhanced phonetizer and stress restoration system and more.

• French and Canadian French: phrasing/chunker have been aligned on the voices’ natural phrasing as well as breathing model, part of speech tagger has been enhanced. New Anais voice.

• Dutch: Part of speech analysis and stress model has been updated.

• German: Improved morphological analysis leading to more accurate pronunciations/phrasing on compound words. German ClaudiaSmile joins our standard portfolio and offers a commercial-smile variation on Claudia’s’ existing voice.

• Regular maintenance and bug correction on all other languages.

Specific for mobile devices

Add voice mode load option (full preprocessing, preprocessing HD and unplugged)

Specific for Servers

WINDOWS

• Registry configuration replaced by file configuration (all binaries components)
• Remove SAPI 4 support (client)

BOTH

• Handling for special HTML/XML meta characters (' " < > &  ) replaced by single character
• Fix all SSML “says-as” tags replaced by single space character except ‘spelling’ tag replaced by Acatts rms tag
• Support URL file in SSML <audio ..> tag

 

New tags to support Acatts based exceptions/abbreviations (dynamic exceptions/abbreviations and files) allowing to load up 10 files at a time.