Spaces:
Runtime error
Runtime error
File size: 1,654 Bytes
1a942eb |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 |
"""Package which defines utility functions for voice conversion."""
import numpy as np
from numpy.typing import NDArray
import ffmpeg
def load_audio(file: str, sr: int) -> NDArray[np.float32]:
"""
Load an audio file into a numpy array with a target sample rate.
A subprocess is launched to decode the given audio file while
down-mixing and resampling as necessary.
Parameters
----------
file : str
Path to the audio file.
sr : int
Target sample rate.
Returns
-------
NDArray[np.float32]
Decoded audio file in numpy array format.
Raises
------
RuntimeError
If the audio file cannot be loaded.
See Also
--------
https://github.com/openai/whisper/blob/main/whisper/audio.py#L26
Notes
-----
Requires the ffmpeg CLI and `typed-ffmpeg` package to be installed.
"""
try:
# NOTE prevent the input path from containing spaces and
# carriage returns at the beginning and end.
file = file.strip(" ").strip('"').strip("\n").strip('"').strip(" ")
out, _ = (
ffmpeg.input(file, threads=0)
.output(
filename="-",
f="f32le",
acodec="pcm_f32le",
ac=1,
ar=sr,
)
.run(
cmd=["ffmpeg", "-nostdin"],
capture_stdout=True,
capture_stderr=True,
)
)
except Exception as e:
err_msg = f"Failed to load audio: {e}"
raise RuntimeError(err_msg) from e
return np.frombuffer(out, np.float32).flatten()
|