asteroid-team/torch-audiomentations
Fast audio data augmentation in PyTorch. Inspired by audiomentations. Useful for deep learning.
PythonMIT
Issues
- 1
- 0
Device Conversion Bug
#174 opened by nuniz - 0
Remove soundfile backend setting
#172 opened by MorenoLaQuatra - 1
- 0
MPS: Simple use of RoomImpulseResponse gives `TypeError: unsupported operand type(s) for *: 'Tensor' and 'Tensor'`
#169 opened by turian - 10
Loading multiple files into a tensor?
#168 opened by asusdisciple - 2
Impulse Response returning only wet audio
#165 opened by Thanatoz-1 - 1
PyPi ColoredNoise doesn't match GitHub
#167 opened by wilke0818 - 2
Why is RMS norm used in AddBackgroundNoise?
#166 opened by jp1924 - 12
- 1
Return list of applied transforms
#164 opened by laplaceon - 13
No fast pitch-shift ratios could be computed for the given sample rate and transpose range.
#151 opened by Crazylov3 - 3
AddColoredNoise: reasoning for f_decay being sampled from uniform distribution?
#160 opened by fujikosu - 0
- 1
BandStopFilter is not useful for environmental sounds at high sample rates
#159 opened by mariusfaiss - 4
- 2
Make BaseWaveformTransform able to pad transform outputs that have a different length
#158 opened by iver56 - 2
Empty ValueError raised during augmentation
#155 opened by rafaelgreca - 0
- 1
RandomCrop does not work on GPU
#153 opened by miccio-dk - 10
- 0
SpliceOut can not be used with `from_yaml`
#144 opened by HLasse - 0
Switch to ObjectDict as default output format
#149 opened by iver56 - 1
Padding transform
#91 opened by iver56 - 0
- 0
Improve Spliceout
#140 opened by iver56 - 1
Adding 'Mix' is giving an error.
#146 opened by devesh-k - 0
Include Mix in the demo script
#148 opened by iver56 - 1
Fix return type inconsistencies
#141 opened by iver56 - 3
Get rid of librosa dependency in favor of torchaudio
#134 opened by hbredin - 3
Implement SpliceOut
#107 opened by iver56 - 3
Implement NoiseGate
#108 opened by iver56 - 2
[Bug] RuntimeError: Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cpu
#137 opened by ahikaml - 4
a multi-gpu bug
#136 opened by DingJuPeng1 - 3
- 3
- 1
bug in adding background noise when noise is short, sr of input is 22050, and sr of noise is 16000
#94 opened by keunwoochoi - 5
- 0
Actual bandwidth in BPF is 2x that of the wording
#118 opened by iver56 - 1
Implement Vocal Tract Length Perturbation
#115 opened by akashrajkn - 4
Microphone style transfer (MicAugment)
#88 opened by iver56 - 1
Pip installs old version
#110 opened by SolomidHero - 2
bump python version to 3.9 (part 2)
#109 opened by SolomidHero - 13
Implement p_mode="per_example" in Compose()
#90 opened by keunwoochoi - 0
The CI should use environment.yml instead of test_requirements for setting up the python env
#106 opened by iver56 - 5
support python3.9
#102 opened by eschmidbauer - 0
Implement time stretch transform
#100 opened by iver56 - 3
Frequency filtering
#98 opened by crimeacs - 0
Band reject transform
#93 opened by iver56 - 0
Document steepness of LPF & HPF
#89 opened by iver56