Multi-Track Crosstalk Reduction Using Spectral Subtraction

Seipel F, Lerch A. Multi-Track Crosstalk Reduction Using Spectral Subtraction. In: AES Convention 144. Audio Engineering Society; 2018. Paper 10013. Available from: https://aes.org/publications/elibrary-page/?id=19409

@inproceedings{Seipel2018_19409,
  author = {Seipel, Fabian and Lerch, Alexander},
  title = {{Multi-Track Crosstalk Reduction Using Spectral Subtraction}},
  booktitle = {AES Convention 144},
  note = {Paper 10013},
  year = {2018},
  month = may,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=19409}
}

TY  - CPAPER
TI  - Multi-Track Crosstalk Reduction Using Spectral Subtraction
AU  - Seipel, Fabian
AU  - Lerch, Alexander
T2  - AES Convention 144
M1  - Paper 10013
PY  - 2018
DA  - 2018/05/06
UR  - https://aes.org/publications/elibrary-page/?id=19409
PB  - Audio Engineering Society
LA  - en
AB  - While many music-related blind source separation methods focus on mono or stereo material, the detection and reduction of crosstalk in multi-track recordings is less researched. Crosstalk or “bleed” of one recorded channel in another is a very common phenomenon in specific genres such as jazz and classical, where all instrumentalists are recorded simultaneously. We present an efficient algorithm that estimates the crosstalk amount in the spectral domain and applies spectral subtraction to remove it. Randomly generated artificial mixtures from various anechoic orchestral source material were employed to develop and evaluate the algorithm, which scores an average SIR-Gain result of 15.14 dB on various datasets with different amounts of simulated crosstalk.
ER  -