OPTIMAL HUMAN-MACHINE CONVERSATIONS USING EMOTION-ENHANCED NATURAL SPEECH USING ARTIFICIAL NEURAL NETWORKS AND REINFORCEMENT LEARNING
A system and method for emotion-enhanced natural speech using artificial neural networks, wherein two artificial neural networks are trained to recognize emotion in text-based content and audio-based content by processing text-based training data and audio-based training data, wherein the audio-based training data corresponds to the text-based training data; and an emotion injection model is created by associating text from output data of one of the artificial neural networks with sounds from output data of a second of the artificial neural networks based on the emotions associated with each.
1 . A system for emotion-enhanced natural speech audio generation using artificial neural networks, comprising:
a computing device comprising a memory and a processor;
a neural network trainer, comprising a first plurality of programming instructions stored in the memory of the computing device which, when operating on the processor of the computing device, causes the computing device to:
train at least two artificial neural networks to recognize emotion in text-based content and audio-based content by processing text-based training data and audio-based training data, the audio-based training data corresponding to the text-based training data; and
construct an emotion injection model by associating text from output data of one of the artificial neural networks with sounds from output data of a second of the artificial neural networks based on the emotions associated with each; and
an automated emotion engine comprising a second plurality of programming instructions stored in the memory of, and operating on the processor of, the computing device, wherein the programming instructions, when operating on the processor, cause the computing device to:
receive text content;
process the text content through one of the artificial neural networks to recognize emotional states in the text content; and
convert the text content to audio content using a text-to-speech engine, modulating the audio content with the modulations of sounds associated with the text from the emotion injection model.
2 . The system of claim 1 , wherein the first and second artificial neural networks are dilated convolutional artificial neural networks.
3 . A method for emotion-enhanced natural speech audio generation using artificial neural networks, comprising the steps of:
training at least two artificial neural networks to recognize emotion in text-based content and audio-based content by processing text-based training data and audio-based training-data, wherein the audio-based training data corresponds to the text-based training data;
constructing an emotion injection model by associating text from output data of one of the artificial neural networks with sounds from output data of a second of the artificial neural networks based on the emotions associated with each;
receiving text content;
processing the text content through one of the artificial neural networks to recognize emotional states in the text content; and
converting the text content to audio content using a text-to-speech engine, and modulating the audio content with the modulations of sounds associated with the text from the emotion injection model.
4 . The method of claim 3 , wherein the first and second artificial neural networks are dilated convolutional artificial neural networks.