High-Fidelity Stereo Vocoder
AssigneeAdobe Inc.
InventorsGe ZHU, Juan-Pablo CACERES CHOMALI, Nicholas J. BRYAN
Life of this patent
Estimating the term…
Abstract
Embodiments are disclosed for music generation. The method may include obtaining a music representation. A high-fidelity vocoder can then generate low-resolution audio data based on the music representation. A bandwidth extension module generates high-resolution monophonic audio data based on the low-resolution audio data and the music representation. A mono-to-stereo module generates high-resolution stereophonic audio data based on the high-resolution monophonic audio data.
Claims20
A method comprising:
- obtaining a music representation;
- generating, by a high-fidelity vocoder, low-resolution audio data based on the music representation;
- generating, by a bandwidth extension module, high-resolution monophonic audio data based on the low-resolution audio data and the music representation;
- and
- generating, by a mono-to-stereo module, high-resolution stereophonic audio data based on the high-resolution monophonic audio data.
A non-transitory computer-readable medium storing executable instructions, which when executed by a processing device, cause the processing device to perform operations comprising:
- obtaining a music representation;
- generating, by a high-fidelity vocoder, low-resolution audio data based on the music representation;
- generating, by a bandwidth extension module, high-resolution monophonic audio data based on the low-resolution audio data and the music representation;
- and
- generating, by a mono-to-stereo module, high-resolution stereophonic audio data based on the high-resolution monophonic audio data.
A system comprising:
- a memory component;
- and
- a processing device coupled to the memory component, the processing device to perform operations comprising:
- receiving a request to restore input music;
- executing one or more of a plurality of generative adversarial network (GAN) based models to progressively increase fidelity of the input music;
- and
- outputting restored music generated by the one or more of the plurality of GAN based models.
Dependencies are inferred from the claim text.