_user_functions Module#

`_user_functions` Module#

waloviz._user_functions.Audio(source: Union[str, PathLike, BinaryIO, ndarray, Tensor, Any, Tuple[Union[ndarray, Tensor, Any], int]], over_curve: Optional[Union[List[Any], Dict[str, Any], ndarray, Tensor, Tuple[Any, Any], Any]] = None, *args: Tuple, over_curve_names: Optional[Union[str, List[str]]] = None, sr: Optional[int] = None, frame_ms: Optional[int] = None, n_fft: Optional[int] = None, hop_ms: Optional[int] = None, hop_length: Optional[int] = None, title: Optional[str] = 'waloviz', embed_title: bool = False, height: Union[int, str] = 'responsive', width: Union[int, str] = 'responsive', aspect_ratio: Optional[float] = None, sizing_mode: Optional[str] = None, sync_legends: bool = False, colorbar: bool = False, cmap: str = 'Inferno', over_curve_colors: Optional[Union[str, List[Optional[str]], Dict[str, str]]] = None, theme: Union[str, Dict[str, Any]] = 'dark_minimal', max_size: int = 10000, download_button: bool = True, freq_label: Optional[str] = 'Hz', over_curve_axes: Optional[Union[str, List[Optional[str]], List[str], Dict[str, str]]] = None, axes_limits: Optional[Dict[str, Tuple[Optional[Union[float, int]], Optional[Union[float, int]]]]] = None, native_player: bool = False, minimal: bool = False, extended: bool = False) → Viewable[source]#

Create an interactive audio player with a spectrogram.

Examples#

import wavloviz as wv

wv.Audio('http://ccrma.stanford.edu/~jos/mp3/pno-cs.mp3')

wav, sr = np.random.randn(8000), 8000
wv.Audio((wav, sr))

wav, sr = torchaudio.load('http://ccrma.stanford.edu/~jos/mp3/pno-cs.mp3')
wv.Audio(wav, sr=sr, over_curve=dict(squared=wav**2, cubed=wav**3))

Parameters#

sourcestr | os.PathLike | IO | (tensorlike, int) | tensorlike: Either an audio file, or an audio tensor\ndarray with a sample rate
over_curvetensorlike | List[tensorlike] | Dict[str, tensorlike] | callable: A single or multiple curves to be displayed over the spectrogram
over_curve_namesstr | List[str]: A list of display names corresponding to the list given in over_curve
srint: The sample rate, when the source does not contain a sample rate, the given sr value is assumed to be the source sample rate, when this value is different than the source sample rate, the source audio is resampled to the specified sr value.
frame_msfloat: Sets the spectrogram frame length to the given amount of milliseconds, default is 100.0.
n_fftint: Sets the n_fft of the spectrogram, overrides the frame_ms value, default is (sr/1000 * frame_ms) .
hop_msfloat: Sets the hop_length of the spectrogram, in milliseconds.
hop_lengthint: Sets the hop_length of the spectrogram, default is n_fft/8
titlestr: A title to be used when saving the plot. If embed_title is True, the title value will be displayed as part of the plot itself. Default is “waloviz”.
embed_titlebool: When True, the title value will be displayed as part of the plot itself, default is False.
heightint | str: The total height of the plot, default is “responsive”, which means the plot will stretch in height to fit.
widthint | str: The total width of the plot, default is “responsive”, which means the plot will stretch in width to fit.
aspect_ratiofloat: The ratio between the width and height, relevant only when either width, height or both are “responsive”, default is 3.5.
sizing_modestr: The panel sizing_mode , can be one of seven values: “stretch_width”, “stretch_height”, “stretch_both”, “scale_width”, “scale_height”, “scale_both”, or “fixed”. Default is “scale_both”.
sync_legendsbool: Whether the legends of both audio channels over curves should be synchronized, default is False
colorbarbool: Whether to display a colorbar for the spectrograms, default is False
cmapstr: The colormap used to display the spectrogram, default is “Inferno”
over_curve_colorsstr | List[str] | Dict[str, str]: Sets the colors to display for each given over_curve , should match the size and structure of the given over_curve value.
themestr | Dict[str, Any]: Sets the visual look and feel of the plot, the value provided must be a bokeh theme, default is “dark_minimal”.
max_sizeint: When the spectrogram or one of the over curves contain many values, the plot’s performance suffers. For that reason max_size limits the amount of displayed values, when the spectrogram or an over curve has more values than the max_size , it is reduced in size by skipping intermediate values, until the size is less than the max_size . Default is 10000.
download_buttonbool: Whether to show the html download button. Defaults to True.
freq_labelstr: The label of the frequency axis (vertical), hides the label when set to None which saves space.
over_curve_axesstr | List[str] | Dict[str, str]: Sets the axis for each given over_curve , should match the size and structure of the given over_curve value. Setting the axis to “Hz” will sync the curve to the frequency axis of the spectrogram.
axes_limitsDict[str, Tuple[float, float]]: Sets default limits for the different axes, the available axes are “x” for the time dimension, “Hz” for the frequency dimension, “y” for the default over curves dimension, and any other axis provided in the over_curve_axes parameter.
native_playerbool: Whether the underlying native audio player should be visible. Default is False
minimalbool: Does nothing when False, when True it overrides some settings to make the player more compact and simple. Default is False.
extendedbool: Does nothing when False, when True it overrides some settings to make the player more descriptive and functional. Default is False.

Returns#

playerpn.viewable.Viewable: An interactive player, can be saved to html with wv.save(player)

Raises#

ValueError: When both minimal=True and extended=True

OR

When no sample-rate was provided

OR

When the wav tensor had more than 2 non squeezable dimensions

OR

When the theme string value was not found in Bokeh

OR

When there are more than 2 positional args

OR

When the provided over_curve was an integer

waloviz._user_functions.extension(mode: str = 'default') → None[source]#

Initializes the notebook extensions for the current IDE.

It is the equivalent of:

hv.extension("bokeh")
pn.extension(comms="default")

When the mode parameter is “colab” those two lines of code will rerun on every wv.Audio call.

Examples#

import wavloviz as wv
wv.extension()
wv.Audio('http://ccrma.stanford.edu/~jos/mp3/pno-cs.mp3')

Parameters#

modestr: Sets the mode of WaloViz, currently the only active mode is “colab”. Default is “default”

waloviz._user_functions.save(source: Union[Viewable, str, PathLike, BinaryIO, ndarray, Tensor, Any, Tuple[Union[ndarray, Tensor, Any], int]], second_arg: Optional[Union[List[Any], Dict[str, Any], ndarray, Tensor, Tuple[Any, Any], Any, str, PathLike, IO]] = None, *args: Tuple, out_file: Optional[Union[str, PathLike, IO]] = None, title: Optional[str] = None, resources: Resources = Resources(mode='inline'), embed: bool = True, **kwargs: Dict[str, Any]) → Union[str, PathLike, IO][source]#

Saves a player to an html file.

Example#

import wavloviz as wv
wv.save('http://ccrma.stanford.edu/~jos/mp3/pno-cs.mp3')

Parameters#

sourcepn.viewable.Viewable | str | os.PathLike | IO | (tensorlike, int) | tensorlike: The player created by wv.Audio , or a source for wv.Audio to create a player with.
out_filestr | os.PathLike | IO: The output file path for the generated html, default is “{title}.html”
titlestr: The title to be used in the generated file name and the html title, if wv.Audio(title="...") was specified, then that value is used, otherwise, the default is “waloviz”.
resourcesbokeh.resources.Resources: The resources for the panel save method, default is INLINE
embedbool: The embed value for the panel save method, default is True

Returns#

out_filestr | os.PathLike | IO: The file that the HTML player content was written into

Raises#

ValueError: When called with more than 2 positional args