# tflibrosa
re-implementation of torch librosa for tensorflow. It is usefull if you want to compute Spectrogram on GPU for faster inference instead of using librosa.
# Installation
> pip install tflibrosa
# Example
To do some inference on single sample, you can use python script in examples/ folder or use as follows:
```
import numpy as np
from tflibrosa import STFT, Spectrogram, LogmelFilterBank
import librosa
import tensorflow as tf
audio = np.random.uniform(0,1 ,(32000 * 5))
print(audio.shape)
sample_rate = 32000
n_fft = 2048
hop_size = 512
window = 'hann'
pad_mode = 'reflect'
mel_bins = 64
ref = 1.0
amin = 1e-10
fmin = 20
fmax = 16000
top_db = 80.0
center = True
dtype=None
spectrogram_extractor = Spectrogram(n_fft=n_fft, hop_length=hop_size,
win_length=n_fft, window=window, center=center, pad_mode=pad_mode,
freeze_parameters=True, dtype="float32")
# Logmel feature extractor
logmel_extractor = LogmelFilterBank(sr=sample_rate, n_fft=n_fft, is_log=True,
n_mels=mel_bins, fmin=fmin, fmax=fmax, ref=ref, amin=amin, top_db=top_db,
freeze_parameters=True, dtype="float32")
spectrogram = spectrogram_extractor(audio[None, :])
mel_spectrogram = logmel_extractor(spectrogram)
print(mel_spectrogram) # (batch size, num_channels, timestamps)
```
# Acknowledgement
- librosa : https://librosa.org/doc/latest/index.html
- torchlibrosa : https://github.com/qiuqiangkong/torchlibrosa
Raw data
{
"_id": null,
"home_page": "https://github.com/Shiro-LK/tflibrosa",
"name": "tflibrosa",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "tflibrosa,tensorflow,librosa",
"author": "Shiro-LK",
"author_email": "shirosaki94@gmail.com",
"download_url": "https://github.com/Shiro-LK/tflibrosa.git",
"platform": null,
"description": "# tflibrosa\nre-implementation of torch librosa for tensorflow. It is usefull if you want to compute Spectrogram on GPU for faster inference instead of using librosa.\n\n# Installation \n\n> pip install tflibrosa\n\n# Example\n\nTo do some inference on single sample, you can use python script in examples/ folder or use as follows:\n\n```\nimport numpy as np \nfrom tflibrosa import STFT, Spectrogram, LogmelFilterBank\nimport librosa\nimport tensorflow as tf \naudio = np.random.uniform(0,1 ,(32000 * 5))\nprint(audio.shape)\n\nsample_rate = 32000\nn_fft = 2048\nhop_size = 512\nwindow = 'hann'\npad_mode = 'reflect'\nmel_bins = 64\nref = 1.0\namin = 1e-10\nfmin = 20\nfmax = 16000 \ntop_db = 80.0\ncenter = True \ndtype=None\n\nspectrogram_extractor = Spectrogram(n_fft=n_fft, hop_length=hop_size, \n win_length=n_fft, window=window, center=center, pad_mode=pad_mode, \n freeze_parameters=True, dtype=\"float32\")\n\n# Logmel feature extractor\nlogmel_extractor = LogmelFilterBank(sr=sample_rate, n_fft=n_fft, is_log=True, \n n_mels=mel_bins, fmin=fmin, fmax=fmax, ref=ref, amin=amin, top_db=top_db, \n freeze_parameters=True, dtype=\"float32\")\n\n\nspectrogram = spectrogram_extractor(audio[None, :])\n\nmel_spectrogram = logmel_extractor(spectrogram)\n\nprint(mel_spectrogram) # (batch size, num_channels, timestamps)\n```\n\n\n# Acknowledgement\n\n- librosa : https://librosa.org/doc/latest/index.html\n- torchlibrosa : https://github.com/qiuqiangkong/torchlibrosa \n",
"bugtrack_url": null,
"license": "MIT License",
"summary": "Re-implementation of some librosa function for tensorflow. Reproduction from torchlibrosa.",
"version": "0.0.2",
"project_urls": {
"Download": "https://github.com/Shiro-LK/tflibrosa.git",
"Homepage": "https://github.com/Shiro-LK/tflibrosa"
},
"split_keywords": [
"tflibrosa",
"tensorflow",
"librosa"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "f4e0a856218c836498b59d1f99c2fb0217cc49f45d5c850b5f4b27b636cbca96",
"md5": "93d7f78e5d1c843a0151b46af2db82b2",
"sha256": "f4739b19c3f3340e561e32a4b14e691fd8ef95fb01b7be98421672f64993cfe6"
},
"downloads": -1,
"filename": "tflibrosa-0.0.2-py3-none-any.whl",
"has_sig": false,
"md5_digest": "93d7f78e5d1c843a0151b46af2db82b2",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 8172,
"upload_time": "2023-08-26T13:45:35",
"upload_time_iso_8601": "2023-08-26T13:45:35.521998Z",
"url": "https://files.pythonhosted.org/packages/f4/e0/a856218c836498b59d1f99c2fb0217cc49f45d5c850b5f4b27b636cbca96/tflibrosa-0.0.2-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-08-26 13:45:35",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "Shiro-LK",
"github_project": "tflibrosa",
"travis_ci": false,
"coveralls": false,
"github_actions": false,
"lcname": "tflibrosa"
}
JSON
