torch-pitch-shift 1.0.15

No project description provided

Torch Pitch Shift

Pitch-shift 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:

• Pitch-shift audio clips quickly using PyTorch (with CUDA support)
• Calculate efficient pitch-shift targets (useful for augmentation, where speed is more important than precise pitch-shifts)

Installation

pip install torch_pitch_shift

Usage

Example

It's super simple:

# import libraries
import torch
from torch_pitch_shift import *

# specify the sample rate
SAMPLE_RATE = 16000
# create a random stereo audio clip (1s long)
audio = torch.rand(
    2,
    SAMPLE_RATE,
    device="cuda" if torch.cuda.is_available() else "cpu"
)
# create the pitch shifter
pitch_shift = PitchShifter()

# for fast shift targets between -1 and +1 octaves
for ratio in get_fast_shifts(SAMPLE_RATE):
    # shift the audio clip
    shifted = pitch_shift(audio, ratio, SAMPLE_RATE)
    print(f"Pitch shift ({ratio}):", shifted)

Check out example.py to see torch_pitch_shift a more detailed example!

Documentation

See the GitHub Wiki Page for detailed documentation!

Contributing

Please feel free to submit issues or pull requests!