A vocal source separation
Project description
Vocal
A vocal source separation
pip install vocal
import librosa
import soundfile as sf
from vocal import get_model, separate_vocal
device = "cuda" # or cpu
audio, sr = librosa.load("audio.wav", sr=44100, mono=False)
model = get_model(device) # download model from HF
audio_data = separate_vocal(model, audio, device, silent=False)
sf.write("vocal.mp3", format="MP3", data=audio_data.T, samplerate=sr)
CLI
Sinlge file
vocali -i audio.mp3 -o output.mp3
Folder
vocali -i audio/ -o output --recursive
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
vocal-0.0.4.tar.gz
(5.4 kB
view details)
Built Distribution
File details
Details for the file vocal-0.0.4.tar.gz.
File metadata
- Download URL: vocal-0.0.4.tar.gz
- Upload date:
- Size: 5.4 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.8.19
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | f7e287cec58c2dacf8c3bed3bbfafd97500871787c54e3d85a7218fd6f3ef7b3 |
|
| MD5 | 0b33caec31e812d6e301576d69845f32 |
|
| BLAKE2b-256 | 5e308d46d88ba2394e221ea3cdb902a1ff16dd0f6bc651646c0cac94c62d0570 |
File details
Details for the file vocal-0.0.4-py2.py3-none-any.whl.
File metadata
- Download URL: vocal-0.0.4-py2.py3-none-any.whl
- Upload date:
- Size: 5.2 kB
- Tags: Python 2, Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/5.1.1 CPython/3.8.19
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 10c6b9c4147ca7326eb52b627a8038c4db9f5737973dddf6378d5b8dffc01f3f |
|
| MD5 | 1c814d70e3aeab859187f31f7e719f2e |
|
| BLAKE2b-256 | e323a1185e2f52cc906733a98625a20aeecb4c04285ea7ceab693852ff0a3540 |
