Time-stretch audio clips quickly with PyTorch (CUDA supported)! Additional utilities for searching efficient transformations are included.

Last update: Jul 07, 2022

Related tags

Deep Learning torch-time-stretch

Overview

Torch Time Stretch

Time-stretch audio clips quickly with PyTorch (CUDA supported)! Additional utilities for searching efficient transformations are included.

View on PyPI / View Documentation

About

This package includes two main features:

Time-stretch audio clips quickly using PyTorch (with CUDA support)
Calculate efficient time-stretch targets (useful for augmentation, where speed is more important than precise time-stretches)

Also check out torch-pitch-shift, a sister project for pitch-shifting.

Installation

pip install torch-time-stretch

Usage

Example

Check out example.py to see torch-time-stretch in action!

Documentation

See the documentation page for detailed documentation!

Contributing

Please feel free to submit issues or pull requests!

Additional code for Stable-baselines3 to load and upload models from the Hub.

Hugging Face x Stable-baselines3 A library to load and upload Stable-baselines3 models from the Hub. Installation With pip Examples [Todo: add colab t

34 Dec 10, 2022

BYOL for Audio: Self-Supervised Learning for General-Purpose Audio Representation

BYOL for Audio: Self-Supervised Learning for General-Purpose Audio Representation This is a demo implementation of BYOL for Audio (BYOL-A), a self-sup

160 Jan 4, 2023

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

21.3k Jan 1, 2023

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

11.4k Feb 13, 2021

Extending JAX with custom C++ and CUDA code

Extending JAX with custom C++ and CUDA code This repository is meant as a tutorial demonstrating the infrastructure required to provide custom ops in

237 Dec 23, 2022

Several simple examples for popular neural network toolkits calling custom CUDA operators.

Neural Network CUDA Example Several simple examples for neural network toolkits (PyTorch, TensorFlow, etc.) calling custom CUDA operators. We provide

798 Jan 1, 2023

Picasso: A CUDA-based Library for Deep Learning over 3D Meshes

The Picasso Library is intended for complex real-world applications with large-scale surfaces, while it also performs impressively on the small-scale applications over synthetic shape manifolds. We have upgraded the point cloud modules of SPH3D-GCN from homogeneous to heterogeneous representations, and included the upgraded modules into this latest work as well. We are happy to announce that the work is accepted to IEEE CVPR2021.

97 Dec 1, 2022

Code for "Learning Structural Edits via Incremental Tree Transformations" (ICLR'21)

Learning Structural Edits via Incremental Tree Transformations Code for "Learning Structural Edits via Incremental Tree Transformations" (ICLR'21) 1.

40 Dec 23, 2022

This Repo is the official CUDA implementation of ICCV 2019 Oral paper for CARAFE: Content-Aware ReAssembly of FEatures

Introduction This Repo is the official CUDA implementation of ICCV 2019 Oral paper for CARAFE: Content-Aware ReAssembly of FEatures. @inproceedings{Wa

42 Jan 7, 2023

Comments

RuntimeError: The size of tensor a (40264) must match the size of tensor b (173) at non-singleton dimension 1

I use same code in https://github.com/KentoNishi/torch-time-stretch/blob/master/example.py but get below error

(librosa) ➜  torch-time-stretch git:(master) ✗ python example.py 
Traceback (most recent call last):
  File "/home/jackie/code/github/torch-time-stretch/example.py", line 48, in <module>
    test_time_stretch_2_up()
  File "/home/jackie/code/github/torch-time-stretch/example.py", line 20, in test_time_stretch_2_up
    up = time_stretch(sample, Fraction(1, 2), SAMPLE_RATE)
  File "/home/jackie/code/github/torch-time-stretch/torch_time_stretch/main.py", line 116, in time_stretch
    output = stretcher(output)
  File "/home/jackie/anaconda3/envs/librosa/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
    return forward_call(*input, **kwargs)
  File "/home/jackie/anaconda3/envs/librosa/lib/python3.9/site-packages/torchaudio/transforms/_transforms.py", line 1059, in forward
    return F.phase_vocoder(complex_specgrams, rate, self.phase_advance)
  File "/home/jackie/anaconda3/envs/librosa/lib/python3.9/site-packages/torchaudio/functional/functional.py", line 743, in phase_vocoder
    phase = angle_1 - angle_0 - phase_advance
RuntimeError: The size of tensor a (40264) must match the size of tensor b (173) at non-singleton dimension 1

opened by Jackiexiao 4

Example ratios are reversed.

Love it, thanks for making this! Tiny thing: In the example test_time_stretch_2_up should use 1/2 as a ratio, not 2/1. test_time_stretch_2_down should use that 2/1 (it's stretching the clip length by 2x).

opened by hdemmer 1

Does it with mono-channel wav files?

my audio clip is in mono 16khz audio, [ 0 0 0 ... 63 100 127], so it will throw

---> 15 down = time_stretch(sample, Fraction(2, 1), SAMPLE_RATE)
     16 wavfile.write(
     17     "./stretched_down_2.wav",
     18     SAMPLE_RATE,
     19     np.swapaxes(down.cpu()[0].numpy(), 0, 0).astype(dtype),
     20 )

File /opt/conda/envs/classify-audio/lib/python3.9/site-packages/torch_time_stretch/main.py:108, in time_stretch(input, stretch, sample_rate, n_fft, hop_length)
    106 if not hop_length:
    107     hop_length = n_fft // 32
--> 108 batch_size, channels, samples = input.shape
    109 # resampler = T.Resample(sample_rate, int(sample_rate / stretch)).to(input.device)
    110 output = input

ValueError: not enough values to unpack (expected 3, got 2)

opened by ti3x 0

Time-stretch audio clips quickly with PyTorch (CUDA supported)! Additional utilities for searching efficient transformations are included.

Related tags

Overview

Torch Time Stretch

About

Installation

Usage

Example

Documentation

Contributing

You might also like...

Additional code for Stable-baselines3 to load and upload models from the Hub.

BYOL for Audio: Self-Supervised Learning for General-Purpose Audio Representation

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

Extending JAX with custom C++ and CUDA code

Several simple examples for popular neural network toolkits calling custom CUDA operators.

Picasso: A CUDA-based Library for Deep Learning over 3D Meshes

Code for "Learning Structural Edits via Incremental Tree Transformations" (ICLR'21)

This Repo is the official CUDA implementation of ICCV 2019 Oral paper for CARAFE: Content-Aware ReAssembly of FEatures

Comments

RuntimeError: The size of tensor a (40264) must match the size of tensor b (173) at non-singleton dimension 1

Example ratios are reversed.

Does it with mono-channel wav files?

Releases(v1.0.3)

v1.0.3(Sep 5, 2022)

v1.0.2(Oct 10, 2021)

v1.0.1(Oct 10, 2021)

v1.0.0(Oct 10, 2021)

Owner

Kento Nishi

A treasure chest for visual recognition powered by PaddlePaddle

PyTorch implementation for Graph Contrastive Learning with Augmentations

PyTorch implementation for ComboGAN

Program your own vulkan.gpuinfo.org query in Python. Used to determine baseline hardware for WebGPU.

Deep Semisupervised Multiview Learning With Increasing Views (IEEE TCYB 2021, PyTorch Code)

Retinal vessel segmentation based on GT-UNet

The pyrelational package offers a flexible workflow to enable active learning with as little change to the models and datasets as possible

PyTorch implementation of NeurIPS 2021 paper: "CoFiNet: Reliable Coarse-to-fine Correspondences for Robust Point Cloud Registration"

Study of human inductive biases in CNNs and Transformers.

A GridMixup augmentation, inspired by GridMask and CutMix

OBBDetection is a oriented object detection library, which is based on MMdetection.

KaziText is a tool for modelling common human errors.

QQ Browser 2021 AI Algorithm Competition Track 1 1st Place Program

ArcaneGAN by Alex Spirin

PyTorch implementation for Partially View-aligned Representation Learning with Noise-robust Contrastive Loss (CVPR 2021)

NVIDIA container runtime

ML powered analytics engine for outlier detection and root cause analysis.

Code for SyncTwin: Treatment Effect Estimation with Longitudinal Outcomes (NeurIPS 2021)

Pytorch code for our paper Beyond ImageNet Attack: Towards Crafting Adversarial Examples for Black-box Domains)

"MST++: Multi-stage Spectral-wise Transformer for Efficient Spectral Reconstruction" (CVPRW 2022) & (Winner of NTIRE 2022 Challenge on Spectral Reconstruction from RGB)