Identifying language in music
AssigneeLemon Inc.
InventorKeunwoo Choi
Life of this patent
Estimating the term…
Abstract
The present disclosure describes techniques for identifying languages associated with music. Training data may be received, wherein the training data comprise information indicative of audio data representative of a plurality of music samples and metadata associated with the plurality of music samples. The training data further comprises information indicating a language corresponding to each of the plurality of music samples. A machine learning model may be trained to identify a language associated with a piece of music by applying the training data to the machine model until the model reaches a predetermined recognition accuracy. A language associated with the piece of music may be determined using the trained machine learning model.
Claims17
A method of recognizing languages in music by a machine learning model, comprising:
- receiving at least one of audio data of a piece of music or metadata associated with the piece of music by the machine learning model trained to recognize the languages in music, wherein the machine learning model is pre-trained using training data, the training data comprise information indicative of audio data representative of a plurality of music samples and metadata associated with the plurality of music samples, and the training data further comprise information indicating a language corresponding to each of the plurality of music samples, wherein the machine learning model is pre-trained to directly associate audio data and metadata of particular pieces of music with different languages, wherein receiving the audio data of the piece of music comprises generating an image representative of the audio data of the piece of music by preprocessing the audio data of the piece of music, and wherein the image representative of the audio data of the piece of music comprises a spectrogram representative of how a frequency of an audio signal of the piece of music varies with time;
- and
- determining a language associated with the piece of music based on the at least one of the audio data of the piece of music or the metadata associated with the piece of music by the machine learning model.
A system of recognizing languages in music, comprising:
- at least one processor;
- and
- at least one memory communicatively coupled to the at least one processor and storing instructions that upon execution by the at least one processor cause the system to perform operations comprising:
- receiving at least one of audio data of a piece of music or metadata associated with the piece of music by a machine learning model trained to recognize the languages in music, wherein the machine learning model is pre-trained using training data, the training data comprise information indicative of audio data representative of a plurality of music samples and metadata associated with the plurality of music samples, and the training data further comprise information indicating a language corresponding to each of the plurality of music samples, wherein the machine learning model is pre-trained to directly associate audio data and metadata of particular pieces of music with different languages, wherein receiving the audio data of the piece of music comprises generating an image representative of the audio data of the piece of music by preprocessing the audio data of the piece of music, and wherein the image representative of the audio data of the piece of music comprises a spectrogram representative of how a frequency of an audio signal of the piece of music varies with time;
- and
- determining a language associated with the piece of music based on the at least one of the audio data of the piece of music or the metadata associated with the piece of music by the machine learning model.
A non-transitory computer-readable storage medium, storing computer-readable instructions that upon execution by a processor cause the processor to implement operations comprising:
- receiving at least one of audio data of a piece of music or metadata associated with the piece of music by a machine learning model trained to recognize languages in music, wherein the machine learning model is pre-trained using training data, the training data comprise information indicative of audio data representative of a plurality of music samples and metadata associated with the plurality of music samples, and the training data further comprise information indicating a language corresponding to each of the plurality of music samples, wherein the machine learning model is pre-trained to directly associate audio data and metadata of particular pieces of music with different languages, wherein receiving the audio data of the piece of music comprises generating an image representative of the audio data of the piece of music by preprocessing the audio data of the piece of music, and wherein the image representative of the audio data of the piece of music comprises a spectrogram representative of how a frequency of an audio signal of the piece of music varies with time;
- and
- determining a language associated with the piece of music based on at least one of the audio data of the piece of music or the metadata associated with the piece of music by the machine learning model.
Dependencies are inferred from the claim text.