transform_to_tensor {torchaudio}

Convert an audio object into a tensor

Description

Converts a numeric vector, as delivered by the backend, into a torch_tensor of shape (channels x samples). If provided by the backend, attributes "channels" and "sample_rate" will be used.

Usage

transform_to_tensor(
  audio,
  out = NULL,
  normalization = TRUE,
  channels_first = TRUE
)

Arguments

audio

(numeric): A numeric vector, as delivered by the backend.

out

(Tensor): An optional output tensor to use instead of creating one. (Default: NULL)

normalization

(bool, float or function): Optional normalization. If boolean TRUE, then output is divided by 2^(bits-1). If bits info is not available it assumes the input is signed 32-bit audio. If numeric, then output is divided by that number. If function, then the output is passed as a parameter to the given function, then the output is divided by the result. (Default: TRUE)

channels_first

(bool): Set channels first or length first in result. (Default: TRUE)

Value

list(Tensor, int), containing
- the audio content, encoded as `[C x L]` or `[L x C]` where L is the number of audio frames and
    C is the number of channels
- the sample rate of the audio (as listed in the metadata of the file)