公开(公告)号：US12198673B2

公开(公告)日：2025-01-14

申请号：US17525814

申请日：2021-11-12

Applicant: LEMON INC.

Inventor： Lamtharn Hantrakul ,  Siyuan Shan ,  Jitong Chen ,  Matthew David Avent ,  David Trevelyan

IPC: G10L13/04 ,  G06N20/00 ,  G10H7/10 ,  G10L13/047 ,  G10L13/08 ,  G10L15/04 ,  G10L19/26

Abstract: The present disclosure describes techniques for differentiable wavetable synthesizer. The techniques comprise extracting features from a dataset of sounds, wherein the features comprise at least timbre embedding; input the features to the first machine learning model, wherein the first machine learning model is configured to extract a set of N×L learnable parameters, N represents a number of wavetables, and L represents a wavetable length; outputting a plurality of wavetables, wherein each of plurality of wavetables comprises a waveform associated with a unique timbre, the plurality of wavetables form a dictionary, and the plurality of wavetables are portable to perform audio-related tasks. Finally, the said wavetables are used to initialize another machine learning model so as to help reduce computational complexity of an audio synthesis obtained as output of the another machine learning model.

公开(公告)号：US20230154451A1

公开(公告)日：2023-05-18

申请号：US17525814

申请日：2021-11-12

Applicant: LEMON INC.

Inventor： Lamtharn HANTRAKUL ,  Siyuan Shan ,  Jitong Chen ,  Matthew David Avent ,  David Trevelyan

IPC: G10L13/04 ,  G06N20/00

CPC classification number: G10L13/04 ,  G06N20/00

Abstract: The present disclosure describes techniques for differentiable wavetable synthesizer. The techniques comprise extracting features from a dataset of sounds, wherein the features comprise at least timbre embedding; input the features to the first machine learning model, wherein the first machine learning model is configured to extract a set of N×L learnable parameters, N represents a number of wavetables, and L represents a wavetable length; outputting a plurality of wavetables, wherein each of plurality of wavetables comprises a waveform associated with a unique timbre, the plurality of wavetables form a dictionary, and the plurality of wavetables are portable to perform audio-related tasks.

公开(公告)号：US12283284B2

公开(公告)日：2025-04-22

申请号：US17748882

申请日：2022-05-19

Applicant: Lemon Inc.

Inventor： Lamtharn Hantrakul ,  David Trevelyan ,  Haonan Chen ,  Matthew David Avent ,  Janne Jayne Harm Renée Spijkervet

IPC: G10L19/00 ,  G10L21/0264 ,  G10L21/04 ,  G10L25/18 ,  G10L25/90

Abstract: Example aspects include techniques for implementing real-time and low-latency synthesis of audio. These techniques may include generating a frame by sampling audio input in increments equal to a buffer size of until a threshold corresponding to a frame size used to train a machine learning (ML) model is reached, detecting feature information within the frame, determining, by the ML model, control information for audio reproduction based on the feature information. In addition, the techniques may include generating filtered noise information by inverting the noise magnitude control information using an overlap and add technique, generating, based on the control information, additive harmonic information by combining a plurality of scaled wavetables, and rendering audio output based on the filtered noise information and the additive harmonic information.