Methods and systems for processing and mixing signals using signal decomposition

1. A system that processes audio signals comprising:

a first microphone that receives a first signal mixture related to a first and a second sound source signal;

a second microphone that receives a second signal mixture related to at least the first and the second sound source signal, wherein the first microphone is a primary microphone for the first and the second sound source signal and is placed close to the first and the second sound source, and wherein the second microphone is an ambient microphone and is placed at a location that is further than the first microphone from the first and the second sound source to capture sound signals coming from a plurality of sources including the first and the second sound source;

wherein a distance between the first and second microphone is capable of being changed;

a signal decomposer that singly operates on the first signal mixture to isolate an intrinsic part of the first sound source signal, wherein an intrinsic part is the tail portion of the first sound source signal;

the signal decomposer also operates on the second signal mixture to isolate the intrinsic part of the first sound source signal;

a storage that saves at least a portion of the intrinsic parts of the first sound source isolated in the first and second signal mixtures; and

a processor that processes only the saved at least a portion of the intrinsic parts of the first sound source signal that meet a similarity measure with respect to one another to output an enhanced first sound source signal, wherein the similarity measure is based on a cosine similarity.

2. The system of claim 1 , wherein one sound source signal is an audio signal from a singing voice, or a stringed instrument, or a wind instrument, or an electric guitar, or a keyboard, or a drum or a musical instrument.

3. The system in claim 1 , wherein the processing of the saved at least a portion of the intrinsic parts of the first sound source signal includes one or more of time-aligning said saved at least a portion of the intrinsic parts, combining said saved at least a portion of the intrinsic parts and selecting said saved at least a portion of the intrinsic parts.

4. The system in claim 3 , wherein said processing and outputting maps the saved at least a portion of the intrinsic parts of the first sound source signal to digital audio workstation tracks.

5. The system in claim 3 , wherein said processing and outputting routes the at least a portion of the intrinsic parts of the first sound source signal one of a portion of said intrinsic parts to at least one mixing channel.

6. The system of claim 1 , wherein the signal decomposer utilizes a signal decomposition method including one or more of non-negative matrix factorization, non-negative tensor factorization, independent component analysis, principal component analysis, singular value decomposition, dependent component analysis, low-complexity coding and decoding, stationary subspace analysis, common spatial pattern, empirical mode decomposition, tensor decomposition, canonical polyadic decomposition, higher-order singular value decomposition, or tucker decomposition.

7. In a system for processing audio signals, comprising a first and second microphone, a method comprising:

receiving, from a first microphone, a first signal mixture related to a first and a second sound source signal;

receiving, from a second microphone, a second signal mixture related to at least the first and the second sound source signal, wherein the first microphone is a primary microphone for the first and the second sound source signal and is placed close to the first and the second sound source, and wherein the second microphone is an ambient microphone and is placed at a location that is further than the first microphone from the first and the second sound source to capture sound signals coming from a plurality of sources including the first and the second sound source;

wherein a distance between the first and second microphone is capable of being changed;

isolating an intrinsic part of the first sound source signal, wherein said isolating is performed by singly operating on the first signal mixture, and wherein an intrinsic part is the onset portion of the first sound source signal;

isolating the intrinsic part of the first sound source signal, wherein said isolating is performed by operating on the second signal mixture;

saving, in a storage, at least a portion of the intrinsic parts of the first sound source isolated in the first and second signal mixtures;

processing only the saved at least a portion of the intrinsic parts of the first sound source that meet a similarity measure with respect to one another to output an enhanced first sound source signal, wherein the similarity measure is based on a cosine similarity.

8. The method of claim 7 , wherein the first sound source signal is an audio signals from a singing voice, or a stringed instrument, or a wind instrument, or an electric guitar, or a keyboard, or a drum or a musical instrument.

9. The method of claim 7 , wherein the processing of the saved at least a portion of the intrinsic parts of the first sound source includes one or more of time-aligning said saved at least a portion of the intrinsic parts, combining said saved at least a portion of the intrinsic parts and selecting said saved at least a portion of the intrinsic parts.

10. The method of claim 9 , wherein said processing and outputting maps the saved at least a portion of the intrinsic parts of the first sound source to digital audio workstation tracks.

11. The method of claim 9 , wherein the processing and outputting routes the saved at least a portion of the intrinsic parts of the first sound source to at least one mixing channel.

12. The method of claim 7 , wherein the isolating utilizes a signal decomposition method including one or more of non-negative matrix factorization, non-negative tensor factorization, independent component analysis, principal component analysis, singular value decomposition, dependent component analysis, low-complexity coding and decoding, stationary subspace analysis, common spatial pattern, empirical mode decomposition, tensor decomposition, canonical polyadic decomposition, higher-order singular value decomposition, or tucker decomposition.

13. A non-transitory computer-readable information storage media having stored thereon instructions, that when executed by a processor, cause to be performed a method comprising:

receiving, from a first microphone, a first signal mixture related to a first and a second sound source signal;

receiving, from a second microphone, a second signal mixture related to at least the first and the second sound source signal, wherein the first microphone is a primary microphone for the first and the second sound source signal and is placed close to the first and the second sound source, and wherein the second microphone is an ambient microphone and is placed at a location that is further than the first microphone from the first and the second sound source to capture sound signals corning from a plurality of sources including the first and the second sound source;

wherein a distance between the first and second microphone is capable of being changed;

isolating an intrinsic part of the first sound source, wherein said isolating is performed by singly operating on the first signal mixture, and wherein an intrinsic part is the onset portion of the first sound source signal;

isolating the intrinsic part of the first sound source, wherein said isolating is performed by operating on the second signal mixture;

saving, in a storage, at least a portion of the intrinsic parts of the first sound source isolated in the first and second signal mixtures;

processing only the saved at least a portion of the intrinsic parts of the first sound source that meet a similarity measure with respect to one another to output an enhanced first sound source signal, wherein the similarity measure is based on a cosine similarity.

14. The media of claim 13 , wherein the first sound source signal is an audio signal from singing voice, or a stringed instrument, or a wind instrument, or an electric guitar, or a keyboard, or a drum or a musical instrument.

15. The media of claim 13 , wherein the processing of the saved at least a portion of the intrinsic parts of the first sound source includes one or more of time-aligning said saved at least a portion of the intrinsic parts, combining said saved at least a portion of the intrinsic parts and selecting said saved at least a portion of the intrinsic parts.

16. The media of claim 15 , wherein said processing and outputting maps the saved at least a portion of the intrinsic parts of the first sound source to digital audio workstation tracks.

17. The media of claim 15 , wherein the processing and outputting routes the saved at least one of a portion of the intrinsic parts of the first sound source to at least one mixing channel.

18. The media of claim 13 , wherein the isolating utilizes a signal decomposition method including one or more of non-negative matrix factorization, non-negative tensor factorization, independent component analysis, principal component analysis, singular value decomposition, dependent component analysis, low-complexity coding and decoding, stationary subspace analysis, common spatial pattern, empirical mode decomposition, tensor decomposition, canonical polyadic decomposition, higher-order singular value decomposition, or tucker decomposition.