10 results about "Music synthesis" patented technology

The invention belongs to the technical field of sleep health, and particularly relates to a federal generation type AI offline personalized sleep-aiding music synthesis system. The system comprises a multi-mode physiological signal offline acquisition and preprocessing module, a federated learning driven offline sleep-music mapping model training module, a generative AI offline sleep-aiding music synthesis module and an offline music library updating and adaptation optimization module. The multi-mode physiological signal offline acquisition and preprocessing module is used for acquiring user sleep physiological data offline and performing local preprocessing to generate a user sleep physiological feature packet; and the federated learning driven offline sleep-music mapping model training module adopts an offline federated mode to train a global sleep-music mapping model. The method does not need real-time networking, is adaptive to portable equipment such as a mobile phone and an intelligent bracelet, can averagely shorten the sleep time, reduces the number of times of awakening at night, is suitable for household insomnia intervention, travel sleep aiding and other scenes, and gives consideration to privacy safety, personalized accuracy and hearing comfort.

Selecting music that complements spoken language, such as announcements, is a time-consuming task. [Solution] The sound source formation system 1 includes a word matching unit 10 that compares related words in the relational data with words in the sound-related information by referring to audio-related information containing multiple words and relational data in which music and related words are related; a keyword selection unit 20 that selects a keyword from the matching words that match through the matching of related words and words in the sound-related information by the word matching unit 10; a music selection unit 30 that selects music related to the keyword as selected music based on the keyword and relational data; a synthesis unit 40 that forms a sound source by synthesizing audio data related to the sound-related information and the selected music; and a range setting unit that sets the range for matching words in the sound-related information. The word matching unit 10 matches words in the sound-related information within the range set by the range setting unit.

Cover music synthesis includes accessing a first digital song specified in a playlist of one or more digital songs, extracting, from the first digital song, first musical parameters characteristic of a music content of the first digital song, receiving a first vocal embedding indicative of a first singer, the first vocal embedding describing speech spoken by or vocals sung by the first singer, and using the first vocal embedding, generating first synthetic vocals for the first digital song using a vocal synthesis model, the first synthetic vocals matching the first musical parameters. The first synthetic vocals can be combined with a first music audio corresponding to the first digital song to generate a first synthetic song.

The utility model relates to a music synthesizer toy for children, which comprises a shell, a control circuit board, a power supply, a master control and volume adjusting switch, a loudspeaker, musical scale contacts, an external lead, a conductive alligator clip, an electronic DJ table, a recording switch, a playing switch, a rhythm sound effect control switch and a multi-element sound effect change-over switch, the control circuit board, the power supply, the loudspeaker and the musical scale contacts are all arranged in the shell, the main control and volume adjusting switch, the external lead, the electronic DJ table, the playing switch, the rhythm sound effect control switch and the multi-sound effect change-over switch are arranged on the outer wall of the shell and are electrically connected with the control circuit board, the external lead is correspondingly connected with the musical scale contacts, and the electronic DJ table is electrically connected with the playing switch. And the rhythm sound effect control switch is correspondingly connected with the conductive alligator clip and comprises a rhythm sound switch and a drumbeat sound switch. The technical scheme has multiple modes and various playing methods, and children can obtain continuous interest when playing, so that the interest of music is improved, and the hand-eye coordination ability, creativity and imagination are trained.

The present disclosure is concerned with A music synthesizer comprising: an input unit having: a note selection input for enabling a user to input a note by selecting the note from a range of notes; a chord type input for enabling the user to input a chord type; and a chord range commencement note input for enabling the user to input a chord range commencement note; and a controller in communication with the input unit, the controller being configured to: generate a plurality of chord notes for a chord, the chord characterized by: the input note as a root note of the chord; the input chord type as a type of the chord; and a chord voicing as a voicing of the chord, the chord voicing comprising one or more chord notes falling within a single octave range commencing with the input chord range commencement note; and output the generated plurality of chord notes to a sound generator.

The invention relates to the technical field of music generation and artificial intelligence crossing, and discloses a music inspiration concrete interaction system which comprises a gesture acquisition module, a data processing module, a melody contour extraction module, an AI phrase complementation module and a music synthesis output module. Generating high-quality music; real-time performance and creativity are balanced; the self-adaptive capability is strong; the system adopts mature computer vision, a sensor technology and a deep learning algorithm, does not need hardware or algorithm support exceeding the prior art level, and has a good industrialization prospect.

A music synthesizer generates an audio signal using an assembly comprising hundreds or thousands of resonators. These resonators may be based on the analysis of any acoustic space, such as acoustic musical instruments, rooms, recording studios, or music halls. A machine learning network is trained to learn features of the music sound. The feature may be whether the sound is melodious. The network produces an audio effect applied to a selected frequency in the spectrum. An input signal or an excitation signal is provided to the network that processes the input by a trained model of the target audio source and configures the resonator components to generate an output audio signal based on the input signal. The network may be expanded to create novel impulse responses to create tones and timbres unique to existing audio sources, which input signals may include musical tones or include human voice inputs.

The invention provides a training method of a music synthesis model, a music synthesis method, equipment and a medium, and relates to the technical field of artificial intelligence, and the training method comprises the steps: obtaining model training data; performing multi-stage model training on the initial music synthesis model based on the model training data to obtain a trained music synthesis model; wherein different model training data are used when the music synthesis model is subjected to model training in different stages, and the multi-stage model training comprises preheating stage model training, generalization stage model training and annealing stage model training. The preheating stage model training is used for optimizing the lyric following capability of the music synthesis model, the generalization stage model training is used for enhancing the generalization capability of the music synthesis model, and the annealing stage model training is used for optimizing the music creation capability of the music synthesis model. By adopting the method and the device, the model can be optimized in the process of training the music synthesis model, so that the music generated by the music synthesis model is more creative.

The invention provides a music synthesis method, electronic equipment, a storage medium and a computer program product, and relates to the technical field of artificial intelligence. Lyric text data and reference music data of to-be-synthesized music are acquired; performing lyric following processing on the lyric text data and the reference music data based on a large language model in the target music synthesis model to obtain a second discrete audio feature, and performing acoustic information reduction processing on the second discrete audio feature based on a stream matching model in the target music synthesis model to obtain an acoustic spectrum feature; and performing audio waveform reduction based on the acoustic spectrum features to obtain synthetic music data corresponding to the lyric text data. By adopting the method, the stability of the music synthesis model for maintaining the music synthesis effect can be improved, so that the availability of the synthesized music and the user experience are ensured.

A musical synthesizer produces an audio signal using a set including hundreds or thousands of resonators. The resonators can be based on analysis of any acoustic space such as an acoustic instrument, room, studio, or concert hall A machine learning network is trained to learn the characteristics of a musical sound. The characteristic may be whether the sound is pleasing to the human ear. The network produces audio effects applied to selected frequencies in the spectrum. An input or excitation signal is provided to the network, which processes the input through a trained model of a target audio source and configures the set of resonators to produce an output audio signal based on the input signal. The network may be expanded to create novel impulse responses creating tones and timbre unique to existing audio sources, the input signal may include musical tones or include vocal inputs.