A lyrics analyzer generates tags and explicitness indicators for a set of tracks. These tags may indicate the genre, mood, occasion, or other features of each track. The lyrics analyzer does so by generating an n-dimensional vector relating to a set of topics extracted from the lyrics and then using those vectors to train a classifier to determine whether each tag applies to each track. The lyrics analyzer may also generate playlists for a user based on a single seed song by comparing the lyrics vector or the lyrics and acoustics vectors of the seed song to other songs to select songs that closely match the seed song. Such a playlist generator may also take into account the tags generated for each track.

Embodiments of the present disclosure provide a harmony processing method and apparatus, a device, and a medium, and the method includes: acquiring a harmonic interval corresponding to a target harmony control in response to a triggering operation on the target harmony control; performing, according to the harmonic interval, sound modification processing on a first sound input originally to obtain a second sound, in which an interval between the first sound and the second sound is the harmonic interval; and generating a target audio according to the first sound and the second sound, wherein the first sound and the second sound are presented as different harmonic parts in the target audio.

Methods and apparatus to classify media based on a pitch-independent timbre attribute from a media signal are disclosed. An example apparatus includes an interface to receive a media signal; a timbre database to store reference pitch-less timbre spectrums; and a processor to: compare a pitch-less timbre spectrum of the media signal to the reference pitch-less timbre spectrums; and classify the media signal based on data corresponding to a reference pitch-less timbre spectrum of the reference pitch-less timbre spectrums that matches the pitch-less timbre spectrum, the classification corresponding to at least one of an instrument or a genre.

Example articles of manufacture and apparatus for producing supplemental information for audio signature data are disclosed herein. An example apparatus includes memory including computer readable instructions. The example apparatus also includes a processor to execute the instructions to at least obtain first audio signature data associated with a first time period of media, obtain first semantic signature data associated with the first time period of the media and second semantic signature data associated with a second time period of the media, and when second audio signature data associated with the second time period of the media is unavailable, identify the media based on the first audio signature data associated with the first time period of media when the second semantic signature data associated with the second time period matches the first semantic signature data associated with the first time period of the media.

It is disclosed a method comprising obtaining a target spectrum, obtaining a set of non-target spectra, the set of non-target spectra comprising one or more non-target spectra, summing the target spectrum and the set of non-target spectra to obtain a mixture spectrum, and training an artificial neural network by using the mixture spectrum as input of the neural network and by using a spectrum which is based on the target spectrum as desired output of the artificial neural network.

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.

Music may be selected to provide accompaniment for a video edit of a video. Characteristics of the music may be determined and used to select the types of visual effects that are applied in the video edit. The characteristics of the music may be extracted from MIDI file/metadata track containing MIDI information for the music.

Methods and systems for automatic audio loop generation from an audio track identify suitable portions of the audio track for generating audio loops. One or more embodiments identify portions of the audio track that include a beginning beat and an ending beat that have similar audio features that provide for seamless transitions when generating the audio loops. One or more embodiments generate scores for the portions based on the similarity of the audio features of the corresponding beginning and ending beats. Additionally, one or more embodiments use the generated scores to determine whether each portion is a suitable audio loop candidate. One or more embodiments then generate one or more audio loops using one or more suitable portions of the audio track.

Methods and apparatus to classify media based on a pitch-independent timbre attribute from a media signal are disclosed. An example apparatus includes an interface to receive a media signal; a timbre database to store reference pitch-less timbre spectrums; and a processor to: compare a pitch-less timbre spectrum of the media signal to the reference pitch-less timbre spectrums; and classify the media signal based on data corresponding to a reference pitch-less timbre spectrum of the reference pitch-less timbre spectrums that matches the pitch-less timbre spectrum, the classification corresponding to at least one of an instrument or a genre.

An audio information processing method and apparatus are provided. The method includes decoding a first audio file to acquire a first audio subfile corresponding to a first sound channel and a second audio subfile corresponding to a second sound channel; extracting first audio data from the first audio subfile; extracting second audio data from the second audio subfile; acquiring a first audio energy value of the first audio data; acquiring a second audio energy value of the second audio data; and determining an attribute of at least one of the first sound channel and the second sound channel based on the first audio energy value and the second audio energy value.