Ultra-anthropomorphic timbre, professional cloning, major upgrade of timbre design models, supporting 40 languages and 19 tone tags

Now, TransDuck has updated a new generation of speech models for ultra-anthropomorphic timbre, accurately restoring true tone details and comprehensively improving timbre similarity.

##New engine features

• Support **40 ** languages (Including: Chinese, Cantonese, English, Spanish, French, Russian, German, Portugal, Arabic, Italian, Japanese, Korean, Indonesian, Vietnam, Turkish, Dutch, Ukraine, Thai, Polish, Romania, Greek, Czech, Finnish, Hindi, Bulgarian, Danish, Hebrew, Malay, Persian, Slovakia, Swedish, Croatian, Filipino, Hungarian, Norway, Slovenia, Catalan, Ninosk, Tamil, Afrika), which can realize cross-language text to speech

• Support **19 ** modal particle tags: (laughs),(sighs),(coughs),(chuckles),(breath),(inhale),(exhale),(gasps),(pant),(pant),(sniffs),(clear-throat) clear your throat,(snorts) snort,(groans) groans,(burps) sneezes,(lip-smacking) smacking,(humming) humming,(hissing) hissing,(emm) um, you can control the audio generation content more finely

• Support pause control. Usage: Add the <#x#> mark to the text, x is the pause duration (unit: seconds), range [0.01, 99.99], and keep at most two decimal places. The text interval time needs to be set between two texts that can be pronunciation, and multiple pause marks cannot be used continuously

##Related to functions

The functions covered by this engine upgrade include:

• "Super-anthropomorphic tone" in system tone
• Sound Clone (Professional Edition) creates timbre
• Sound design creates sound