This folder contains the code and scripts used for
- Automating the experiment
- Processing the recordings into a usable form
- Compress/Uncompress the samples
Here's a short description for each file here:
| File | Description |
|---|---|
calibration.py |
Routines for the calibration of microphones and sources locations |
compress.sh |
Compress a bunch of wav files to tta format using ffmpeg |
compute_tdoa.py |
Compute the time difference of arrivals between a reference mic and the others |
deconvolution.py |
Routine for the Wiener deconvolution |
ffmpeg_audio.py |
Reads an audio file with FFMPEG |
gen_sweeps.py |
Generates linear and exponential sweeps |
run_calibration.py |
Produce a calibration file for the microphones and sources locations |
run_deconvolution.py |
Produce all the impulse responses for the dataset |
segment.py |
Segment the raw recordings into individual samples |
stitch_samples.py |
Stitch the different audio samples before playback |
uncompress.sh |
Uncompress tar-ed compressed audio to uncompressed audio (directly) |
All this code was written by Robin Scheibler and is under the MIT License.