| | |
|---|
| | | |
|---|
| | | def __init__( |
|---|
| | | self, |
|---|
| | | fs: Union[int, str] = 16000, |
|---|
| | | fs: int = 16000, |
|---|
| | | n_fft: int = 512, |
|---|
| | | win_length: int = None, |
|---|
| | | hop_length: int = 128, |
|---|
| | |
|---|
| | | **kwargs, |
|---|
| | | ): |
|---|
| | | super().__init__() |
|---|
| | | if isinstance(fs, str): |
|---|
| | | fs = humanfriendly.parse_size(fs) |
|---|
| | | |
|---|
| | | # Deepcopy (In general, dict shouldn't be used as default arg) |
|---|
| | | frontend_conf = copy.deepcopy(frontend_conf) |
|---|
| | | self.hop_length = hop_length |
|---|
| | | self.fs = fs |
|---|
| | | |
|---|
| | | if apply_stft: |
|---|
| | | self.stft = Stft( |
|---|
| | |
|---|
| | | return self.n_mels |
|---|
| | | |
|---|
| | | def forward( |
|---|
| | | self, input: torch.Tensor, input_lengths: torch.Tensor |
|---|
| | | self, input: torch.Tensor, input_lengths: Union[torch.Tensor, list] |
|---|
| | | ) -> Tuple[torch.Tensor, torch.Tensor]: |
|---|
| | | if isinstance(input_lengths, list): |
|---|
| | | input_lengths = torch.tensor(input_lengths) |
|---|
| | | if input.dtype == torch.float64: |
|---|
| | | input = input.float() |
|---|
| | | # 1. Domain-conversion: e.g. Stft: time -> time-freq |
|---|
| | | if self.stft is not None: |
|---|
| | | input_stft, feats_lens = self._compute_stft(input, input_lengths) |
|---|
| | |
|---|
| | | |
|---|
| | | def __init__( |
|---|
| | | self, |
|---|
| | | fs: Union[int, str] = 16000, |
|---|
| | | fs: int = 16000, |
|---|
| | | n_fft: int = 512, |
|---|
| | | win_length: int = None, |
|---|
| | | hop_length: int = None, |
|---|
| | |
|---|
| | | mc: bool = True |
|---|
| | | ): |
|---|
| | | super().__init__() |
|---|
| | | if isinstance(fs, str): |
|---|
| | | fs = humanfriendly.parse_size(fs) |
|---|
| | | |
|---|
| | | # Deepcopy (In general, dict shouldn't be used as default arg) |
|---|
| | | frontend_conf = copy.deepcopy(frontend_conf) |
|---|
| | | if win_length is None and hop_length is None: |
|---|
