The present disclosure provides a method for chorus mixing, an apparatus, an electronic device and storage media. The method includes converting a main vocal audio signal and a chorus audio signal into signals in frequency domain, respectively, wherein the chorus audio signal comprises main vocal audio played by a speaker; determining a delay between the main vocal audio signal and the chorus audio signal based on a frequency-domain signal of the main vocal audio signal and a frequency-domain signal of the main vocal audio played by the speaker included in a frequency-domain signal of the chorus audio signal; aligning the chorus audio signal with the main vocal audio signal based on the determined delay; performing an echo cancellation on the aligned chorus audio signal; and mixing audio of the main vocal audio signal and the echo-canceled chorus audio signal.

A performance agent training method realized by at least one computer includes observing a first performance of a musical piece by a performer, generating, by a performance agent, performance data of a second performance to be performed in parallel with the first performance, outputting the performance data such that the second performance is performed in parallel with the first performance of the performer, acquiring a degree of satisfaction of the performer with respect to the second performance performed based on the output performance data, and training the performance agent by reinforcement learning, using the degree of satisfaction as a reward.

An electronic device comprising circuitry configured to separate by audio source separation a first audio input signal into a first vocal signal and an accompaniment, and to transpose an audio output signal by a transposition value based on a pitch ratio, wherein the pitch ratio is based on comparing a first pitch range of the first vocal signal and a second pitch range of the second vocal signal.

A recommendation information provision device includes at least one processor that is configured to acquire a result of scoring relating to a user's singing of a musical piece for each section in time; acquire pitch information representing pitches of notes configuring the musical piece in the section; build a learning model predicting a result of scoring relating to the user's singing of a musical piece from the pitch information using the result of scoring and the pitch information as training data; acquire a result of scoring relating to a user's singing of a target musical piece by inputting the pitch information to the learning model while changing pitches of notes into a plurality of types; and output a setting detail of the pitches of the notes as the recommendation information on the basis of results of scoring for the pitch information of the plurality of types.

The present disclosure relates to a sound source file structure, to output lyrics as audible sounds right before melodies corresponding to the lyrics start, to help a user to remind the lyrics based on accompaniment for a song after the accompaniment starts to be provided, and to help the user to sing based on correct lyrics corresponding to the melodies. The sound source file structure may include one or more backing sound source layers in which backing sounds based on beats and rhythms are placed, a melody sound source layer in which melody notes corresponding to lyrics based on beats and rhythms and a rest section corresponding to a rest are placed, and a lyric voice source layer in which a lyric voice is placed at a position corresponding to a rest section.

User interface techniques provide user vocalists with mechanisms for solo audiovisual capture and for seeding subsequent performances by other users (e.g., joiners). Audiovisual capture may be against a full-length work or seed spanning much or all of a pre-existing audio (or audiovisual) work and in some cases may mix, to seed further contributions of one or more joiners, a user's captured media content for at least some portions of the audio (or audiovisual) work. A short seed or short segment may span less than all (and in some cases, much less than all) of the audio (or audiovisual) work. For example, a verse, chorus, refrain, hook or other limited “chunk” of an audio (or audiovisual) work may constitute a short seed or short segment. Computational techniques are described that allow a system to automatically identify suitable short seeds or short segments. After audiovisual capture against the short seed or short segment, a resulting, solo or group, full-length or short-form performance may be posted, livestreamed, or otherwise disseminated in a social network.

A BLUETOOTH communications system includes a true wireless stereo (TWS) BLUETOOTH headset and a terminal device, where the TWS Bluetooth headset includes a first earbud and a second earbud. The terminal device controls the first earbud to collect a sound signal and the second earbud to play a sound signal. When an audio application on the terminal device is started, the first earbud collects a first sound signal, performs sound effect processing on the first sound signal to obtain a second sound signal, and sends the second sound signal to the second earbud. The terminal device sends accompaniment audio to the second earbud, and the second earbud performs audio mixing processing on the accompaniment audio and the second sound signal for playing.

An electronic device having circuitry configured to perform source separation on an audio signal to obtain a separated source and a residual signal, to perform feature extraction on the separated source to obtain one or more processing parameters, and to perform audio processing on a captured audio signal based on the one or more processing parameters to obtain an adjusted separated source.

The present invention provides a chordophone including a body configured to have a top plate, a bottom plate, and a side plate connected to the top plate and the bottom plate, which comprises: a cut-out plate formed by cutting the bottom plate, and configured to have a grille formed on a portion penetrated of the cut-out plate; a bracket formed in a shape corresponding to the cut-out plate and configured to have a base plate coupled to an inner side of the cut-out plate; and a speaker unit configured to have a speaker installed in the bracket, a control unit for controlling an output of the speaker, and a communication module for allowing a sound source output from the speaker to be received from an external user terminal, in which the grille is formed in a size and shape corresponding to a diameter of a side on which a sound of the speaker is output, and a through-hole is formed penetrated on the base plate at a corresponding position to the grille and with a corresponding shape to a shape of the sound output side of the speaker.

The present disclosure provides method and apparatus for providing personalized songs in automated chatting. A message may be received in a chat flow. Personalized lyrics of a user may be generated based at least on a personal language model of the user in response to the message. A personalized song may be generated based on the personalized lyrics. The personalized song may be provided in the chat flow.