Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

GCC-PHAT Re-Imagined - A U-Net Filter for Audio TDOA Peak-Selection

Gulin, Jens LU orcid and Åström, Kalle LU orcid (2024) 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing p.8806-8810
Abstract
Time-difference-of-arrival (TDOA) estimation from GCC-PHAT is not always as straight forward as finding the maximum peak. This work views the GCC output as an image, with time on the vertical axis and TDOA horizontally, to explore if image-to-image machine learning methods can make a more robust filter. The Structure from Sound Database provides audio recorded with a distributed microphone setup and a moving sound source. The audio was fed to GCC-PHAT without pre-processing, and images were produced for batch processing. The ground truth, the direct-path TDOA, shows a continuous curve through time. The GCC output image has a similar curve, but obscured by noise and not at all times texturally different from the multi-path components. The... (More)
Time-difference-of-arrival (TDOA) estimation from GCC-PHAT is not always as straight forward as finding the maximum peak. This work views the GCC output as an image, with time on the vertical axis and TDOA horizontally, to explore if image-to-image machine learning methods can make a more robust filter. The Structure from Sound Database provides audio recorded with a distributed microphone setup and a moving sound source. The audio was fed to GCC-PHAT without pre-processing, and images were produced for batch processing. The ground truth, the direct-path TDOA, shows a continuous curve through time. The GCC output image has a similar curve, but obscured by noise and not at all times texturally different from the multi-path components. The main approach tested is binary semantic segmentation with a U-Net. A challenge is the extreme class imbalance within the image. Preliminary results indicate that the method is valid to detect curves, yet more work is needed to single out the direct path TDOA with confidence. (Less)
Please use this url to cite or link to this publication:
author
and
organization
alternative title
GCC-PHAT ombildad - Ett U-Net filter för urval av Audio TDOA toppar
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
keywords
Time-difference-of-arrival, Semantic segmentation, curve detection, noise reduction, U-Net, Generalized Cross-Correlation
host publication
ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
pages
5 pages
publisher
IEEE - Institute of Electrical and Electronics Engineers Inc.
conference name
2024 IEEE International Conference on Acoustics, Speech, and Signal Processing
conference location
Seoul, Korea, Republic of
conference dates
2024-04-14 - 2024-04-19
ISBN
979-8-3503-4486-8
979-8-3503-4485-1
DOI
10.1109/ICASSP48485.2024.10447558
language
English
LU publication?
yes
additional info
“© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.”
id
e85c159b-3969-479d-b475-30b3e3e7ab01
date added to LUP
2024-04-29 15:25:59
date last changed
2024-05-03 02:54:41
@inproceedings{e85c159b-3969-479d-b475-30b3e3e7ab01,
  abstract     = {{Time-difference-of-arrival (TDOA) estimation from GCC-PHAT is not always as straight forward as finding the maximum peak. This work views the GCC output as an image, with time on the vertical axis and TDOA horizontally, to explore if image-to-image machine learning methods can make a more robust filter. The Structure from Sound Database provides audio recorded with a distributed microphone setup and a moving sound source. The audio was fed to GCC-PHAT without pre-processing, and images were produced for batch processing. The ground truth, the direct-path TDOA, shows a continuous curve through time. The GCC output image has a similar curve, but obscured by noise and not at all times texturally different from the multi-path components. The main approach tested is binary semantic segmentation with a U-Net. A challenge is the extreme class imbalance within the image. Preliminary results indicate that the method is valid to detect curves, yet more work is needed to single out the direct path TDOA with confidence.}},
  author       = {{Gulin, Jens and Åström, Kalle}},
  booktitle    = {{ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)}},
  isbn         = {{979-8-3503-4486-8}},
  keywords     = {{Time-difference-of-arrival; Semantic segmentation; curve detection; noise reduction; U-Net; Generalized Cross-Correlation}},
  language     = {{eng}},
  month        = {{03}},
  pages        = {{8806--8810}},
  publisher    = {{IEEE - Institute of Electrical and Electronics Engineers Inc.}},
  title        = {{GCC-PHAT Re-Imagined - A U-Net Filter for Audio TDOA Peak-Selection}},
  url          = {{https://lup.lub.lu.se/search/files/181844482/ICASSP24_GCC_Reimagined_Gulin_str_m_.pdf}},
  doi          = {{10.1109/ICASSP48485.2024.10447558}},
  year         = {{2024}},
}