Abstract: A system is configurable to: after receiving first user input, (i) initiate playback of selected audio content at one or more playback components and (ii) initiate audio recording at one or more recording components to obtain recorded audio content, the recorded audio content being recorded at least partially during the playback of the selected audio content; process the selected audio content and the recorded audio content as inputs to an alignment module to generate a temporal offset value, wherein the alignment module is configured to generate the temporal offset value by correlating feature frames of the selected audio content with feature frames of the recorded audio content; and after receiving second user input, initiate synchronized playback of the selected audio content and the recorded audio content, wherein the recorded audio content is synchronized with the selected audio content using the temporal offset value.

Abstract: A system is configurable to generate a set of lyric labels by processing audio content using a first machine learning model, wherein each lyric label of the set of lyric labels is associated with a respective lyric label timestamp; generate a set of chord labels by processing the audio content using a second machine learning model, wherein each chord label of the set of chord labels is associated with a respective chord label timestamp; determine a matching lyric label from the set of lyric labels for at least one chord label of the set of chord labels based on temporal similarity of the respective chord label timestamp of the at least one chord label to the respective lyric label timestamp of the matching lyric label; and generate a display comprising lyric content and chord content that visually correlate the matching lyric label with the at least one chord label.

Abstract: A hardware peripheral is configurable to: (i) receive at least a part of an input audio signal from a first source; (ii) process the part of the input audio signal using a stem separation module to generate stem-separated audio, the stem-separated audio comprising a plurality of audio stems corresponding to different audio sources represented in the part of the input audio signal; and (iii) trigger playback of one or more of the plurality of audio stems by sending the one or more of the plurality of audio stems to a playback device that is in communication with the hardware peripheral such that, following the hardware peripheral receiving the part of the input audio signal, the playback of the one or more of the plurality of audio stems begins at the playback device in real-time or near-real-time.

Abstract: A method comprises: (i) presenting a user interface frontend that displays a control for activating a system-level stem separation mode; and (ii) after receiving user input activating the system-level stem separation mode, and when a system generates a system output audio signal based on audio associated with one or more applications executing on the system: (a) refraining from sending the system output audio signal to one or more playback devices over a system audio output signal path; (b) identifying a set of one or more segments of the system output audio signal; (c) processing the set of one or more segments using a selected stem separation module to generate a set of stem-separated segments comprising a plurality of audio stems; and (d) sending at least one audio stem from the plurality of audio stems to the one or more playback devices via the system audio output signal path.

Abstract: A system is configurable to: (i) cause presentation of an audio processing workflow generation interface comprising: (a) a module tray comprising a plurality of audio processing modules and (b) a workflow canvas configured for placement and display of audio processing modules selected for inclusion in an audio processing workflow; (ii) receive user input selecting a set of audio processing modules from the plurality of audio processing modules of the module tray; (iii) receive user input directed to the workflow canvas in association with the set of audio processing modules; and (iv) place and display a set of representations of the set of audio processing modules on the workflow canvas, wherein each representation of the set of representations includes an input node and an output node and is associated with a respective audio processing module of the set of audio processing modules.

Abstract: A device is configurable to cause playback of an input audio signal and, after detecting user input directed to activating a stem separation mode: (i) cause cessation of playback of the input audio signal; and (ii) until a stop condition is satisfied, iteratively: (a) identify a set of one or more segments of the input audio signal; (b) process the set of one or more segments using a stem separation module to generate a set of stem-separated segments, the set of stem-separated segments comprising a plurality of audio stems corresponding to different audio sources represented in the set of one or more segments; and (c) cause playback of at least one audio stem from the plurality of audio stems.