Creates a waveform visualization from audio tensor data.

Expects a tensor with shape [channels, samples] where:

• channels: number of audio channels
• samples: number of audio samples

Options

• :width - Width of the plot in pixels. Defaults to 800
• :height - Height of the plot in pixels. Defaults to 200
• :color - Color of the waveform. Defaults to "steelblue"
• :background - Background color. Defaults to "white"

Returns a VegaLite specification that can be rendered.