AudioLoader¶

class modusa.io.audio_loader.AudioLoader[source]¶

Bases: ModusaIO

Loads audio from various sources like filepath, YouTube, etc.

Note

All from_ methods return AudioSignal instance.

static from_youtube(url: str, sr: int | None = None) → AudioSignal[source]¶

Loads audio from youtube url using YoutubeDownloader, AudioConverter and librosa.

from modusa.io import AudioSignalLoader

# From youtube
audio_signal = AudioSignalLoader.from_youtube(
        url="https://www.youtube.com/watch?v=lIpw9-Y_N0g",
        sr=None
)

Parameters:

url (str) – Link to the YouTube video.
sr (int) – Sampling rate to load the audio in.

Returns:

Audio signal instance with loaded audio content from YouTube.

Return type:

AudioSignal

static from_fp(fp: str | Path, sr: int | None = None) → AudioSignal[source]¶

Loads audio from a filepath using librosa.

from modusa.io import AudioSignalLoader

# From file
audio_signal = AudioSignalLoader.from_fp(
        fp="path/to/audio.wav",
        sr=None
)

Parameters:

fp (str | Path) – Local filepath of the audio.
sr (int | None) – Sampling rate to load the audio in.

Returns:

Audio signal instance with loaded audio content from filepath.

Return type:

AudioSignal

static from_array(y: ndarray, sr: int) → AudioSignal[source]¶

Loads audio with a given sampling rate.

from modusa.io import AudioSignalLoader
import numpy as np

# From numpy array
audio_signal = AudioSignalLoader.from_array_with_sr(
        x=np.random.random((100, )),
        sr = 100 # Automatically generates time index
)

Parameters:

y (np.ndarray) – Data of the audio signal.
sr (int) – Sampling rate of the audio signal.

Returns:

Audio signal instance with loaded audio content from sampling rate.

Return type:

AudioSignal

static from_list(y: list, t: list) → AudioSignal[source]¶

Loads AudioSignal instance from python list.

from modusa.io import AudioSignalLoader

# From list
audio_signal = AudioSignalLoader.from_list(
        y=[1, 2, 3, 2, 3],
        t = [0.1, 0.2, 0.3, 0.4, 0.5]
)

Parameters:

y (list) – Data of the audio signal.
t (np.ndarray | None) – Corresponding time stamps of the audio signal.

Returns:

Audio signal instance with loaded audio content from python list.

Return type:

AudioSignal