torchaudio.functional.rtf_power¶
- torchaudio.functional.rtf_power(psd_s: Tensor, psd_n: Tensor, reference_channel: Union[int, Tensor], n_iter: int = 3, diagonal_loading: bool = True, diag_eps: float = 1e-07) Tensor[source]¶
Estimate the relative transfer function (RTF) or the steering vector by the power method.
- Parameters:
psd_s (torch.Tensor) – The complex-valued power spectral density (PSD) matrix of target speech. Tensor with dimensions (…, freq, channel, channel).
psd_n (torch.Tensor) – The complex-valued power spectral density (PSD) matrix of noise. Tensor with dimensions (…, freq, channel, channel).
reference_channel (int or torch.Tensor) – Specifies the reference channel. If the dtype is
int, it represents the reference channel index. If the dtype istorch.Tensor, its shape is (…, channel), where thechanneldimension is one-hot.diagonal_loading (bool, optional) – If
True, enables applying diagonal loading topsd_n. (Default:True)diag_eps (float, optional) – The coefficient multiplied to the identity matrix for diagonal loading. It is only effective when
diagonal_loadingis set toTrue. (Default:1e-7)
- Returns:
The estimated complex-valued RTF of target speech. Tensor of dimension (…, freq, channel).
- Return type:
- Tutorials using
rtf_power:
