Skip to content
master
Switch branches/tags
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

WELCOME to the dECHORATE dataset

⚠️ This work is under review: Be very careful here!

dECHORATE: A calibrated dataset for echo-aware audio signal processing

The dEchorate dataset is a new database of measured multichannel Room Impulse Responses (RIRs) including annotations of early echo timings and 3D positions of microphones, real sources and image sourcesunder different wall configurations in a cuboid room.
These data provide a tool for benchmarking recentmethods inecho-awarespeech enhancement, room geometry estimation, RIR estimation, acoustic echoretrieval, microphone calibration, echo labeling and reflectors estimation.
The database is accompanied withsoftware utilities to easily access, manipulate and visualize the data as well as baseline methods forecho-related tasks.

Keywords: Echo-aware signal processing; Acoustic echoes; Room impulse response; Audio database; AcousticEcho Retrieval; Spatial Filtering; Room Geometry Estimation; Microphone arrays.

You can find a detailed explanation of the dEchorate dataset at: dEchorate: a Calibrated Room Impulse Response Database for Echo-aware Signal Processing

dEchorate has three main elements:

  • the data (available on Zenodo);
  • the code for working with dEchorate;
  • the code for reproducing the paper.

Contents

News:

  • 03 May 2020: v0.0.1 dEchorate project is alive

Get the Data

The data is available at Zenodo. Please, follow that link to download (part of) the data.

The dataset is available in multiple ways:

  • Annotations/labels/metadata: csv file that can be used pandas (Python)
  • Only RIRs: numpy matrix n_samples x n_mics x n_srcs x n_rooms (~ 6 GB)
  • Only Speech: numpy matrix n_samples x n_mics x n_srcs x n_rooms (~ 6 GB)
  • Raw data: an hdf5 file (~ 75 GB) containig all the raw recording data (chirps, speech/noise sources, babble noise, room tone)

Examples

Citing dEchorate

If you are using dEchorate and you want to cite us, please use

@preprint{di2021dechorate,
    title={dEchorate: a Calibrated Room Impulse Response Database for Echo-aware Signal Processing},
    author={Di Carlo, Diego and Tandeitnik, Pinchas and Foy, C{\'e}dric and Deleforge, Antoine and Bertin, Nancy and Gannot, Sharon},
    journal={arXiv preprint arXiv:2104.13168},
    year={2021}
  }

About

Da - ECHO - RetrievAl - daTasEt

Resources

License

Releases

No releases published

Packages

No packages published