Method and system providing service based on user voice

A method for providing a service based on a user's voice comprising:

• extracting by a voice acquisition unit a voice of a first user;
• generating by an artificial neural network unit text information or voice waveform information based on the voice of the first user;
• analyzing by the artificial neural network unit a disposition of the first user based on the text information and the voice waveform information, and then selecting by the artificial neural network unit a second user corresponding to the disposition of the first user based on the analysis result;
• providing by the artificial neural network unit the first user with a conversation connection service with the second user;
• and
• acquiring by the artificial neural network unit information on a change in an emotional state of the first user based on conversation information between the first user and the second user, and re-selecting by the artificial neural network unit the second user corresponding to the disposition of the first user based on the acquired information on the change in the emotional state of the first user,
• wherein analyzing the disposition of the first user and selecting the second user comprises:
• performing deep-learning by the artificial neural network unit based on a speech emotion recognition model based on CNN (Convolutional Neural Network) by applying at least one of user voice information and user profile information as an input value and applying a user's emotional information and disposition information as an output value to analyze the disposition of the first user and select the second user, and
• wherein the speech emotion recognition model based on CNN applies a structure implemented by applying multinomial logistic regression after connecting the output value of each neural network instead of a Softmax layer to a concatenate layer.