.. _op_ai_onnx_STFT:

STFT
====

- **Domain**: ``ai.onnx``
- **Since version**: 17

Computes the Short-time Fourier Transform of the signal.

**Inputs**

- **signal** (*T1*): Input tensor representing a real or complex valued signal. For real input, the following shape is expected: [batch_size][signal_length][1]. For complex input, the following shape is expected: [batch_size][signal_length][2], where [batch_size][signal_length][0] represents the real component and [batch_size][signal_length][1] represents the imaginary component of the signal.
- **frame_step** (*T2*): The number of samples to step between successive DFTs.
- **window** (*T1*): A tensor representing the window that will be slid over the signal.The window must have rank 1 with shape: [window_shape]. It's an optional value.
- **frame_length** (*T2*): A scalar representing the size of the DFT. It's an optional value.

**Outputs**

- **output** (*T1*): The Short-time Fourier Transform of the signals.If onesided is 1, the output has the shape: [batch_size][frames][dft_unique_bins][2], where dft_unique_bins is frame_length // 2 + 1 (the unique components of the DFT) If onesided is 0, the output has the shape: [batch_size][frames][frame_length][2], where frame_length is the length of the DFT.

**Attributes**

- **onesided** (*int*): If onesided is 1, only values for w in [0, 1, 2, ..., floor(n_fft/2) + 1] are returned because the real-to-complex Fourier transform satisfies the conjugate symmetry, i.e., X[m, w] = X[m,w]=X[m,n_fft-w]\*. Note if the input or window tensors are complex, then onesided output is not possible. Enabling onesided with real inputs performs a Real-valued fast Fourier transform (RFFT).When invoked with real or complex valued input, the default value is 1. Values can be 0 or 1.

**Type Constraints**

- **T1**: Constrain signal and output to float tensors.
  Allowed types: tensor(bfloat16), tensor(double), tensor(float), tensor(float16).
- **T2**: Constrain scalar length types to int64_t.
  Allowed types: tensor(int32), tensor(int64).

Examples
--------

**test_cc_stft**

.. code-block:: text

    Node:
      STFT(signal, frame_step, "", frame_length) -> (output)
      Attributes:
        onesided = 1

.. code-block:: text

    Inputs:
      signal: shape=(1, 16, 1), dtype=float32
        [[[1.],
          [2.],
          [3.],
          [4.],
          [5.],
          [6.],
          [7.],
          [8.],
          [8.],
          [7.],
          [6.],
          [5.],
          [4.],
          [3.],
          [2.],
          [1.]]]
      frame_step: shape=(), dtype=int64
        4
      frame_length: shape=(), dtype=int64
        8

    Outputs:
      output: shape=(1, 3, 5, 2), dtype=float32
        [[[[ 3.6000000e+01,  0.0000000e+00],
           [-4.0000000e+00,  9.6568546e+00],
           [-4.0000000e+00,  4.0000000e+00],
           [-4.0000000e+00,  1.6568543e+00],
           [-4.0000000e+00, -3.9188699e-15]],
        
          [[ 5.2000000e+01,  0.0000000e+00],
           [-5.8284273e+00, -2.4142137e+00],
           [-2.1431318e-15, -8.8817842e-16],
           [-1.7157288e-01, -4.1421357e-01],
           [ 0.0000000e+00, -2.2043643e-15]],
        
          [[ 3.6000000e+01,  0.0000000e+00],
           [ 4.0000000e+00, -9.6568546e+00],
           [ 4.0000000e+00, -4.0000000e+00],
           [ 4.0000000e+00, -1.6568543e+00],
           [ 4.0000000e+00, -4.8985874e-16]]]]

**test_cc_stft_complex_batched**

.. code-block:: text

    Node:
      STFT(signal, frame_step, "", frame_length) -> (output)
      Attributes:
        onesided = 0

.. code-block:: text

    Inputs:
      signal: shape=(2, 128, 2), dtype=float32
        [[[ 1.,  0.],
          [ 1.,  0.],
          [ 1.,  0.],
          ...,
          [ 1.,  0.],
          [ 1.,  0.],
          [ 1.,  0.]],
        
         [[99.,  0.],
          [99.,  0.],
          [99.,  0.],
          ...,
          [99.,  0.],
          [99.,  0.],
          [99.,  0.]]]
      frame_step: shape=(), dtype=int64
        16
      frame_length: shape=(), dtype=int64
        32

    Outputs:
      output: shape=(2, 7, 32, 2), dtype=float32
        [[[[ 3.2000000e+01,  0.0000000e+00],
           [-2.3314684e-15, -2.2482016e-15],
           [-3.3306691e-16,  0.0000000e+00],
           ...,
           [-3.2973624e-14, -3.2751579e-14],
           [ 6.1395333e-14, -3.2418512e-14],
           [ 4.9960036e-14, -2.8810287e-14]],
        
          [[ 3.2000000e+01,  0.0000000e+00],
           [-2.3314684e-15, -2.2482016e-15],
           [-3.3306691e-16,  0.0000000e+00],
           ...,
           [-3.2973624e-14, -3.2751579e-14],
           [ 6.1395333e-14, -3.2418512e-14],
           [ 4.9960036e-14, -2.8810287e-14]],
        
          [[ 3.2000000e+01,  0.0000000e+00],
           [-2.3314684e-15, -2.2482016e-15],
           [-3.3306691e-16,  0.0000000e+00],
           ...,
           [-3.2973624e-14, -3.2751579e-14],
           [ 6.1395333e-14, -3.2418512e-14],
           [ 4.9960036e-14, -2.8810287e-14]],
        
          ...,
        
          [[ 3.2000000e+01,  0.0000000e+00],
           [-2.3314684e-15, -2.2482016e-15],
           [-3.3306691e-16,  0.0000000e+00],
           ...,
           [-3.2973624e-14, -3.2751579e-14],
           [ 6.1395333e-14, -3.2418512e-14],
           [ 4.9960036e-14, -2.8810287e-14]],
        
          [[ 3.2000000e+01,  0.0000000e+00],
           [-2.3314684e-15, -2.2482016e-15],
           [-3.3306691e-16,  0.0000000e+00],
           ...,
           [-3.2973624e-14, -3.2751579e-14],
           [ 6.1395333e-14, -3.2418512e-14],
           [ 4.9960036e-14, -2.8810287e-14]],
        
          [[ 3.2000000e+01,  0.0000000e+00],
           [-2.3314684e-15, -2.2482016e-15],
           [-3.3306691e-16,  0.0000000e+00],
           ...,
           [-3.2973624e-14, -3.2751579e-14],
           [ 6.1395333e-14, -3.2418512e-14],
           [ 4.9960036e-14, -2.8810287e-14]]],
        
        
         [[[ 3.1680000e+03,  0.0000000e+00],
           [ 2.8421709e-14, -9.5923269e-14],
           [-8.5265128e-14, -8.5265128e-14],
           ...,
           [-3.2827074e-12, -3.2329694e-12],
           [ 6.1533001e-12, -3.1405989e-12],
           [ 5.1727511e-12, -3.0588865e-12]],
        
          [[ 3.1680000e+03,  0.0000000e+00],
           [ 2.8421709e-14, -9.5923269e-14],
           [-8.5265128e-14, -8.5265128e-14],
           ...,
           [-3.2827074e-12, -3.2329694e-12],
           [ 6.1533001e-12, -3.1405989e-12],
           [ 5.1727511e-12, -3.0588865e-12]],
        
          [[ 3.1680000e+03,  0.0000000e+00],
           [ 2.8421709e-14, -9.5923269e-14],
           [-8.5265128e-14, -8.5265128e-14],
           ...,
           [-3.2827074e-12, -3.2329694e-12],
           [ 6.1533001e-12, -3.1405989e-12],
           [ 5.1727511e-12, -3.0588865e-12]],
        
          ...,
        
          [[ 3.1680000e+03,  0.0000000e+00],
           [ 2.8421709e-14, -9.5923269e-14],
           [-8.5265128e-14, -8.5265128e-14],
           ...,
           [-3.2827074e-12, -3.2329694e-12],
           [ 6.1533001e-12, -3.1405989e-12],
           [ 5.1727511e-12, -3.0588865e-12]],
        
          [[ 3.1680000e+03,  0.0000000e+00],
           [ 2.8421709e-14, -9.5923269e-14],
           [-8.5265128e-14, -8.5265128e-14],
           ...,
           [-3.2827074e-12, -3.2329694e-12],
           [ 6.1533001e-12, -3.1405989e-12],
           [ 5.1727511e-12, -3.0588865e-12]],
        
          [[ 3.1680000e+03,  0.0000000e+00],
           [ 2.8421709e-14, -9.5923269e-14],
           [-8.5265128e-14, -8.5265128e-14],
           ...,
           [-3.2827074e-12, -3.2329694e-12],
           [ 6.1533001e-12, -3.1405989e-12],
           [ 5.1727511e-12, -3.0588865e-12]]]]

**test_cc_stft_twosided**

.. code-block:: text

    Node:
      STFT(signal, frame_step, "", frame_length) -> (output)
      Attributes:
        onesided = 0

.. code-block:: text

    Inputs:
      signal: shape=(1, 16, 1), dtype=float32
        [[[1.],
          [2.],
          [3.],
          [4.],
          [5.],
          [6.],
          [7.],
          [8.],
          [8.],
          [7.],
          [6.],
          [5.],
          [4.],
          [3.],
          [2.],
          [1.]]]
      frame_step: shape=(), dtype=int64
        4
      frame_length: shape=(), dtype=int64
        8

    Outputs:
      output: shape=(1, 3, 8, 2), dtype=float32
        [[[[ 3.6000000e+01,  0.0000000e+00],
           [-4.0000000e+00,  9.6568546e+00],
           [-4.0000000e+00,  4.0000000e+00],
           [-4.0000000e+00,  1.6568543e+00],
           [-4.0000000e+00, -3.9188699e-15],
           [-4.0000000e+00, -1.6568543e+00],
           [-4.0000000e+00, -4.0000000e+00],
           [-4.0000000e+00, -9.6568546e+00]],
        
          [[ 5.2000000e+01,  0.0000000e+00],
           [-5.8284273e+00, -2.4142137e+00],
           [-2.1431318e-15, -8.8817842e-16],
           [-1.7157288e-01, -4.1421357e-01],
           [ 0.0000000e+00, -2.2043643e-15],
           [-1.7157288e-01,  4.1421357e-01],
           [-1.7551422e-14, -2.6645353e-15],
           [-5.8284273e+00,  2.4142137e+00]],
        
          [[ 3.6000000e+01,  0.0000000e+00],
           [ 4.0000000e+00, -9.6568546e+00],
           [ 4.0000000e+00, -4.0000000e+00],
           [ 4.0000000e+00, -1.6568543e+00],
           [ 4.0000000e+00, -4.8985874e-16],
           [ 4.0000000e+00,  1.6568543e+00],
           [ 4.0000000e+00,  4.0000000e+00],
           [ 4.0000000e+00,  9.6568546e+00]]]]

**test_cc_stft_with_window**

.. code-block:: text

    Node:
      STFT(signal, frame_step, window, frame_length) -> (output)
      Attributes:
        onesided = 1

.. code-block:: text

    Inputs:
      signal: shape=(1, 16, 1), dtype=float32
        [[[1.],
          [2.],
          [3.],
          [4.],
          [5.],
          [6.],
          [7.],
          [8.],
          [8.],
          [7.],
          [6.],
          [5.],
          [4.],
          [3.],
          [2.],
          [1.]]]
      frame_step: shape=(), dtype=int64
        4
      window: shape=(8,), dtype=float32
        [1., 1., 1., 1., 1., 1., 1., 1.]
      frame_length: shape=(), dtype=int64
        8

    Outputs:
      output: shape=(1, 3, 5, 2), dtype=float32
        [[[[ 3.6000000e+01,  0.0000000e+00],
           [-4.0000000e+00,  9.6568546e+00],
           [-4.0000000e+00,  4.0000000e+00],
           [-4.0000000e+00,  1.6568543e+00],
           [-4.0000000e+00, -3.9188699e-15]],
        
          [[ 5.2000000e+01,  0.0000000e+00],
           [-5.8284273e+00, -2.4142137e+00],
           [-2.1431318e-15, -8.8817842e-16],
           [-1.7157288e-01, -4.1421357e-01],
           [ 0.0000000e+00, -2.2043643e-15]],
        
          [[ 3.6000000e+01,  0.0000000e+00],
           [ 4.0000000e+00, -9.6568546e+00],
           [ 4.0000000e+00, -4.0000000e+00],
           [ 4.0000000e+00, -1.6568543e+00],
           [ 4.0000000e+00, -4.8985874e-16]]]]