Prediction device
An object is to improve accuracy of prediction of a result of scoring relating to user's singing of a new musical piece. A prediction device is a device that acquires a result of scoring relating to a user's singing of a musical piece in the past for each section of the musical piece in time, acquires pitch information representing pitches of notes configuring the musical piece and aligned in a time series in the section, builds a learning model predicting a result of scoring relating to the user's singing of a musical piece from the pitch information using the result of scoring and the pitch information as training data, and acquires a result of scoring relating to the user's singing of a new musical piece on the basis of an output of the learning model by inputting the pitch information about the new musical piece to the learning model.
1 . A prediction device predicting a result of scoring, the prediction device comprising:
at least one processor,
the at least one processor being configured to:
acquire a result of scoring relating to a user's singing of a musical piece in the past for each of a plurality of sections of the musical piece, wherein each section is defined by a time interval;
acquire pitch information representing pitches of notes included in the musical piece and aligned in a time series in each of the plurality of section;
build a learning model predicting a result of scoring relating to the user's singing of a new musical piece using the result of scoring and the pitch information of the musical piece sung in the past as training data; and
acquire a result of scoring relating to the user's singing of the new musical piece on the basis of an output of the learning model by inputting pitch information about the new musical piece to the learning model.
2 . The prediction device according to claim 1 , wherein the at least one processor builds the learning model such that the output of the learning model becomes close to a result of scoring for each section included in the training data, wherein the learning model outputs a result of scoring for each of the new musical piece corresponding to the pitch information.
3 . The prediction device according to claim 2 , wherein the at least one processor acquires the result of scoring relating to the user's singing of the new musical piece by averaging results of scoring for each sections of the new musical piece that are outputs of the learning model.
4 . The prediction device according to claim 1 , wherein the learning model further receives identification information of the user as an input.
5 . The prediction device according to claim 1 , wherein the at least one processor repeatedly acquires results of scoring relating to the user's singing of a plurality of new musical pieces, selects one of the new musical pieces to be recommended to the user on the basis of the results of scoring, and outputs selection information identifying the selected musical piece.
6 . A method, implemented by at least one processor of a prediction device predicting a result of scoring, the method comprising:
acquiring a result of scoring relating to a user's singing of a musical piece in the past for each of a plurality of sections of the musical piece, wherein each section is defined by a time interval;
acquiring pitch information representing pitches of notes included in the musical piece and aligned in a time series in each of the plurality of sections;
building a learning model predicting a result of scoring relating to the user's singing of a new musical piece using the result of scoring and the pitch information of the musical piece sung in the past as training data; and
acquiring a result of scoring relating to the user's singing of the new musical piece on the basis of an output of the learning model by inputting pitch information about the new musical piece to the learning model.