A system and computer-implemented method for generating music content includes a music notation data store having a collection of notation data files and an audio data store having a collection of audio data files, each data file in the notation and audio data stores including associated music characteristic metadata. One or more computer processor is arranged to receive user music preference inputs from a user interface and to search the notation and audio data stores to identify a plurality of data files corresponding to one or more user preference input. The processor randomly selects at least one notation file and at least one audio file from the identified notation and audio files and generates a music instance file by combining the selected notation and audio files for playback to the user.

A method for automating audio mashup production is disclosed. First, two or more audio files are received. Based on two or more audio files, two or more stem audio files and reference metadata associated with the two or more audio files are retrieved from a server. Each of the two or more stem audio files includes at least one of an instrument portion or a vocal portion that are included in the two or more audio files. After retrieval of the two or more stem audio files and the reference metadata, at least some musical parameters associated with segments of the two or more stem audio files are adjusted. Thereafter, the two or more stem audio files or adjusted segments of the two or more stem audio files can be combined into a single audio file. The single audio file is output to a user device.

A music composition system for composing music segments comprises: a computer interface comprising at least one external input for receiving from an external device a request for a musical composition; a controller configured to determine based on a request received at the external input a plurality of musical parts for the musical composition; and a composition engine configured to generate, for each of the determined musical parts, at least one musical segment in digital musical notation format, the musical segments configured to cooperate musically when performed simultaneously. The computer interface comprises at least one external output configured to output a response to the request, the request comprising or indicating each of the musical segments in digital musical notation format for rendering into audio data at the external device.

The disclosed embodiments provide a system that generates sound primitives to facilitate sound recognition. First, the system performs a feature-detection operation on sound samples to detect a set of sound features, wherein each sound feature comprises a measurable characteristic of a window of consecutive sound samples. Next, the system creates feature vectors from coefficients generated by the feature-detection operation, wherein each feature vector comprises a set of coefficients for sound features detected in a window. The system then performs a clustering operation on the feature vectors to produce feature-vector clusters, wherein each feature-vector cluster comprises a set of feature vectors that are proximate to each other in a feature-vector space that contains the feature vectors. After the clustering operation, the system defines a set of sound primitives, wherein each sound primitive is associated with a feature-vector cluster. Finally, the system associates semantic labels with the set of sound primitives.

Methods, systems and apparatus to characterize audio using transchromagrams are disclosed. An example method includes generating, by executing one or more instructions on a processor, a set of transition matrices based on a plurality of time frames of the audio data, each of the plurality of transition matrices generated based on a different pair of time frames in the plurality of time frames, and indicating probabilities that anterior musical notes in an anterior time frame of the pair transition to posterior musical notes in a posterior time frame of the pair, generating, by executing one or more instructions on a processor, a data structure representing how the audio data changes statistically between the plurality of time frames based on the set of transition matrices, and causing, by executing one or more instructions on a processor, a database to store the data structure within metadata that describes the audio data.

An auditory augmentation system includes a database with a multiplicity of audio sections and associated metadata for digital audio files. Each audio section is mapped to a contextual theme, each contextual theme being mapped to an audio section having an entry point and an exit point. The entry and exit points support seamless splice or fade transitions between different audio sections. A processing system couples to the database along with an input; the input is in the form of temporally-varying events data that defines a temporal input. The processing system resolves the temporal input into one or more of a plurality of categorized contextual themes, correlates the categorized contextual themes with metadata associated with selected audio sections relevant to the one or more categorized contextual themes, and splices or fades together selected audio sections, and generates, as an output, a media product in which transitions between audio sections are seamless.

Described herein is a system for automatically mixing music. An ordering component receives information regarding a set of songs, obtains pre-analyzed metadata about songs in the set of songs, and, based upon the pre-analyzed metadata orders the songs into a mix based upon an analysis of beats-per-minute and song key. A cue point component, based upon a user preference, selects one or more cue points from the pre-analyzed metadata for transitions into and/or out of the songs in the mix. A transition component, based upon the user preference and information derived from the pre-analyzed metadata for each section of a particular song, configures transitions between songs in the mix. The system can store metadata about the mix for playback.

One embodiment provides a method, including: receiving, at an information handling device, a user input to play media files associated with a media file type from a playlist comprising a plurality of media files; analyzing, using a processor, the plurality of media files to identify at least one media file not associated with the media file type; disregarding, at least temporarily, based on the analyzing, the at least one media file; and providing, based on the disregarding, output of a media file from the playlist other than the at least temporarily disregarded at least one media file. Other aspects are described and claimed.

Methods and systems are provided for detecting and cataloging qualities in music. While both the data volume and heterogeneity of the digital music content is huge, it has become increasingly important and convenient to build a recommendation or search system to facilitate surfacing these content to the user or consumer community. Embodiments use deep convolutional neural network to imitate how human brain processes hierarchical structures in the auditory signals, such as music, speech, etc., at various timescales. This approach can be used to discover the latent factor models of the music based upon acoustic hyper-images that are extracted from the raw audio waves of music. These latent embeddings can be used either as features to feed to subsequent models, such as collaborative filtering, or to build similarity metrics between songs, or to classify music based on the labels for training such as genre, mood, sentiment, etc.

A content server generates sponsored audio including procedurally generated background music. The content server obtains reference music features describing musicological characteristics of reference songs as well as sponsored audio information received from a third-party. The content server determines music generation parameters based on the sponsored audio information and based on a music model mapping the reference music features to music generation parameters. The music model may incorporate machine learning techniques to improve the mapping. The content server generates background music by using the determined music generation parameters as input to a procedural music algorithm, and generates sponsored audio by combining the generated background music concurrently with an audio voiceover obtained from the sponsored audio information. The sponsored audio is provided to a client device, which presents the sponsored audio to a user.