The team of researchers from the University of East Anglia, Caltech, Carnegie Mellon University and Disney have presented a method to to animate speech in real-time.
The team of researchers from the University of East Anglia, Caltech, Carnegie Mellon University and Disney have presented a method to to animate speech in real-time. The new way automatically incorporates new dialogues in much less time with a lot less effort.
The team recorded over eight hours of audio and video of a speaker reciting more than 2500 different sentences. The speaker’s face was tracked and the data was used to create a reference face for an animation model. Then they used special to transcribe the speech sounds. This whole process trained a neural network to animate a reference face, frame-by-frame, based on phonemes.
Training the AI is said to take only a couple of hours, letting specialists use speech from any speaker with any accent and even in different languages. The method is also capable of dealing with the singing.
Abstract
We introduce a simple and effective deep learning approach to automatically generate natural looking speech animation that synchronizes to input speech. Our approach uses a sliding window predictor that learns arbitrary nonlinear mappings from phoneme label input sequences to mouth movements in a way that accurately captures natural motion and visual coarticulation effects. Our deep learning approach enjoys several attractive properties: it runs in real-time, requires minimal parameter tuning, generalizes well to novel input speech sequences, is easily edited to create stylized and emotional speech, and is compatible with existing animation retargeting approaches. One important focus of our work is to develop an effective approach for speech animation that can be easily integrated into existing production pipelines. We provide a detailed description of our end-to-end approach, including machine learning design decisions. Generalized speech animation results are demonstrated over a wide range of animation clips on a variety of characters and voices, including singing and foreign language input. Our approach can also generate on-demand speech animation in real-time from user speech input.