Cereproc’s Technology Enhances Sophia’s Voice Through Human-Centered AI

Hanson Robotics and Cereproc Extra Collaborate on Human-Centered AI to Make stronger Sophia the Robot’s Narrate

Sophia the Robot of Hanson Robotics has a particular enlighten acknowledged around the sector. Her talking and singing enlighten used to be developed as a collaboration between Hanson Robotics and CereProc, a textual state material-to-speech solutions firm known for synthesizing character-rich voices using deep finding out. As a results of this ongoing collaboration, Sophia famously learned to explain, and performed a duet with Jimmy Fallon on the Tonight Roar.

Since then, Hanson Robotics and CereProc hang persisted to give a increase to Sophia’s enlighten capabilities and attain synthetic speech expertise. In their most up-to-date collaboration, Sophia has gained the flexibility to mimic the inflection, stress, and tone of a human enlighten. The utilization of a neural network model, Sophia can become a human’s enlighten into her have in right-time, all while declaring the desired intonation and stress. In other words, a particular person can verbalize into a microphone to expose Sophia study how to speak a brand recent phrase, and Sophia’s enlighten will doubtless be ready to mimic it. Equally, Sophia can study to explain recent musical passages by precisely copying pre-recorded audio, or by mimicking a particular person singing in right-time, all in her have odd enlighten.

These recent abilities will allow Sophia to declare extra dramatic and captivating performances. Having extra true regulate over Sophia’s inflection will allow her to declare better punchlines, sound extra pure, and better raise the hidden that manner in the aid of her words. As an illustration, folks generally use upward inflection to indicate a question, even when giving a declarative commentary. To expose sarcasm, folks generally use inverse pitch obtrusion, the keep they decrease their pitch for a stressed out phrase in space of elevating it. With these recent capabilities, Sophia will doubtless be ready to harness the vitality of subtle vocal cues widely passe by humans to raise deeper layers of that manner.

Broader Implications

This recent expertise permits a human speaker to use their very have enlighten to finely regulate any synthetic enlighten. This is also beautiful to a differ of industries including choices in healthcare, buyer service agents, social robotics, filmmakers, sport builders, and immersive training companies. As an illustration, creating enlighten synthesis for an actor who plays one of many characters and later having a varied particular person talking into the microphone and to hang the output being a textual state material-to-speech enlighten of the actor. This expertise would possibly maybe well be very beautiful to the film exchange, as this is able to clear up scheduling disorders basically based totally on the actor’s availability and keep on prolonged term production bills.

So not only does this recent expertise hang the doubtless to revolutionize the textual state material to speech exchange in film, sport pattern, and social robotics, you would possibly perchance well well simply one day soon expertise the overwhelming pleasure of having your very have sarcastic singing robotic.

Cereproc and Hanson Robotics Joint Press Delivery HR FINAL 2020 04 24