Computer-based systems, devices, and methods for automatically generating aesthetic chord progressions and key modulations in musical compositions are described. Known harmonic relationships are expanded upon to produce a much richer set of harmonic transition probability models compared to conventional music theory, and these models are leveraged by a computer-based musical composition system to generate new musical compositions and variations of existing musical compositions. Techniques for enabling a computer-based musical composition system to automatically determine when to introduce a key modulation, what key to module to, and what chord progression(s) to use within the new key are all described.

An approach is provided in which an information handling system configures a reinforcement learning model based inspiration selections received from a user. The information handling system performs training iterations using the configured reinforcement learning model, which generates multiple actions and multiple rewards corresponding to multiple actions. The information handling system determines that the multiple rewards reach an empirical threshold and, in turn, generates a musical composition based on the multiple actions.

Performance information of main music is sequentially acquired, and an accent position of the music is determined. An automatic accompaniment is progressed based on accompaniment pattern data. Upon determination that the current time point coincides with the accent position, an accompaniment event whose tone generation timing arrives within a predetermined time range following the current time point is extracted from the accompaniment pattern data, the tone generation timing of the extracted accompaniment event is shifted to the current time point, and then, accompaniment data is created based on the accompaniment event having the tone generation timing thus shifted. If there is no accompaniment event whose tone generation timing arrives at the current time point or within the predetermined time range following the current time point, automatic accompaniment data with the current time point set as its tone generation timing is additionally created.

A method and system for analyzing patterns in the relationships of notes of an input piece of music. The method comprises generating a set of the most frequently occurring note pitches in ascending pitch order that matches an interval pattern, and detecting out-of-key pitches that lie outside of this interval pattern. One or more potential key sequence bifurcations are identified which represent a list of possible key sequences according to forwards and backwards analysis. By finding patterns of repetition in the chordal sequences that may be generated according to these key sequence bifurcations, a key sequence that allows the most frequently recurring chord sequences may be chosen. Chord sequences may be analyzed by using ghost chords, temporary harmonic structures that are created, updated and finalized over time according to a combination of essential and inessential note fragments. The method further comprises identifying non-harmony pitches according to the analyzed chord sequence.

An adaptive music playback system is disclosed. The system includes a composition system that receives information corresponding to user activity levels. The composition system determines target musical criteria corresponding to the user activity levels and modifies the composition of a song in response to changes in user activity.

Methods and system for music generation. One method comprises defining a phrase structure and a metrical layout, and generating a melody based on the phrase structure and the metrical layout using a probabilistic model of contour-sequences in a machine learning model. The probabilistic model includes a plurality of production rules determined by the machine learning model trained on a dataset of hierarchical analyses, and the contour-sequences defining directional patterns between musical notes extracted from the dataset of hierarchical analyses.

An adaptive music playback system is disclosed. The system includes a composition system that receives information corresponding to user activity levels. The composition system modifies the composition of a song in response to changes in user activity. The modifications are made according to a set of composition rules to facilitate smooth musical transitions

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.

The present embodiments relate to a method and technical equipment for audio processing. The method comprises receiving a user input; determining an audio based on the user input; determining rules for adapting the audio based on the user input; receiving data from at least one data source; adapting the audio according to the received data from the at least one data source and according to the rules; generating a data stream comprising at least parameters on how the audio is adapted.

A wireless record booth capable of producing a live video and/or audio stream as if made in a vintage record booth. A concatenated video stream is formed by a record booth sessions recorded in each of a plurality of networked listening/watching record booths. Video and audio filters emulate vintage record booth images and sounds. Record booth sessions are streamed live for other networked record booths to listen and/or watch, and passively and actively rated to dynamically create a concatenated video stream of the most preferred record booth sessions.