View source for Lemon Inc. (20240404494). IMPLEMENTING AUTOMATIC MUSIC AUDIO TRANSCRIPTION

=IMPLEMENTING AUTOMATIC MUSIC AUDIO TRANSCRIPTION=


==Organization Name==

[[:Category:Lemon Inc.|Lemon Inc.]]


[[Category:Lemon Inc.]]

==Inventor(s)==

[[:Category:Wei Tsung Lu of Los Angeles CA (US)|Wei Tsung Lu of Los Angeles CA (US)]][[Category:Wei Tsung Lu of Los Angeles CA (US)]]

[[:Category:Ju-Chiang Wang of Los Angeles CA (US)|Ju-Chiang Wang of Los Angeles CA (US)]][[Category:Ju-Chiang Wang of Los Angeles CA (US)]]

[[:Category:Yun-Ning Hung of Culver City CA (US)|Yun-Ning Hung of Culver City CA (US)]][[Category:Yun-Ning Hung of Culver City CA (US)]]

==IMPLEMENTING AUTOMATIC MUSIC AUDIO TRANSCRIPTION==


This abstract first appeared for US patent application 20240404494 titled 'IMPLEMENTING AUTOMATIC MUSIC AUDIO TRANSCRIPTION




==Original Abstract Submitted==

the present disclosure describes techniques for implementing automatic music audio transcription. a deep neural network model may be configured. the deep neural network model comprises a spectral cross-attention sub-model configured to project a spectral representation of each time step t, denoted as st, into a set of latent arrays at the time step t, denoted as �, h representing an h-th iteration. the deep neutral network model comprises a plurality of latent transformers configured to perform self-attention on the set of latent arrays �. the deep neural network model further comprises a set of temporal transformers configured to enable communications between any pairs of latent arrays �at different time steps. training data may be augmented by randomly mixing a plurality of types of datasets comprising a vocal dataset and an instrument dataset. the deep neural network model may be trained using the augmented training data.



[[Category:G10H1/00]]
[[Category:CPC_G10H1/0066]]