-
Notifications
You must be signed in to change notification settings - Fork 387
/
stoi.py
102 lines (84 loc) · 4.49 KB
/
stoi.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
# Copyright The PyTorch Lightning team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import numpy as np
import torch
from torch import Tensor
from torchmetrics.utilities.checks import _check_same_shape
from torchmetrics.utilities.imports import _PYSTOI_AVAILABLE
if _PYSTOI_AVAILABLE:
from pystoi import stoi as stoi_backend
else:
stoi_backend = None
__doctest_skip__ = ["short_time_objective_intelligibility"]
def short_time_objective_intelligibility(
preds: Tensor, target: Tensor, fs: int, extended: bool = False, keep_same_device: bool = False
) -> Tensor:
r"""STOI (Short-Time Objective Intelligibility, see [2,3]), a wrapper for the pystoi package [1].
Note that input will be moved to `cpu` to perform the metric calculation.
Intelligibility measure which is highly correlated with the intelligibility of degraded speech signals, e.g., due
to additive noise, single/multi-channel noise reduction, binary masking and vocoded speech as in CI simulations.
The STOI-measure is intrusive, i.e., a function of the clean and degraded speech signals. STOI may be a good
alternative to the speech intelligibility index (SII) or the speech transmission index (STI), when you are
interested in the effect of nonlinear processing to noisy speech, e.g., noise reduction, binary masking algorithms,
on speech intelligibility. Description taken from `Cees Taal's website <http://www.ceestaal.nl/code/>`_.
.. note:: using this metrics requires you to have ``pystoi`` install. Either install as ``pip install
torchmetrics[audio]`` or ``pip install pystoi``
Args:
preds: shape ``[..., time]``
target: shape ``[..., time]``
fs: sampling frequency (Hz)
extended: whether to use the extended STOI described in [4]
keep_same_device: whether to move the stoi value to the device of preds
Returns:
stoi value of shape [...]
Raises:
ModuleNotFoundError:
If ``pystoi`` package is not installed
Example:
>>> from torchmetrics.functional.audio.stoi import short_time_objective_intelligibility
>>> import torch
>>> g = torch.manual_seed(1)
>>> preds = torch.randn(8000)
>>> target = torch.randn(8000)
>>> short_time_objective_intelligibility(preds, target, 8000).float()
tensor(-0.0100)
References:
[1] https://github.com/mpariente/pystoi
[2] C.H.Taal, R.C.Hendriks, R.Heusdens, J.Jensen 'A Short-Time Objective Intelligibility Measure for
Time-Frequency Weighted Noisy Speech', ICASSP 2010, Texas, Dallas.
[3] C.H.Taal, R.C.Hendriks, R.Heusdens, J.Jensen 'An Algorithm for Intelligibility Prediction of
Time-Frequency Weighted Noisy Speech', IEEE Transactions on Audio, Speech, and Language Processing, 2011.
[4] J. Jensen and C. H. Taal, 'An Algorithm for Predicting the Intelligibility of Speech Masked by Modulated
Noise Maskers', IEEE Transactions on Audio, Speech and Language Processing, 2016.
"""
if not _PYSTOI_AVAILABLE:
raise ModuleNotFoundError(
"ShortTimeObjectiveIntelligibility metric requires that `pystoi` is installed."
" Either install as `pip install torchmetrics[audio]` or `pip install pystoi`."
)
_check_same_shape(preds, target)
if len(preds.shape) == 1:
stoi_val_np = stoi_backend(target.detach().cpu().numpy(), preds.detach().cpu().numpy(), fs, extended)
stoi_val = torch.tensor(stoi_val_np)
else:
preds_np = preds.reshape(-1, preds.shape[-1]).detach().cpu().numpy()
target_np = target.reshape(-1, preds.shape[-1]).detach().cpu().numpy()
stoi_val_np = np.empty(shape=(preds_np.shape[0]))
for b in range(preds_np.shape[0]):
stoi_val_np[b] = stoi_backend(target_np[b, :], preds_np[b, :], fs, extended)
stoi_val = torch.from_numpy(stoi_val_np)
stoi_val = stoi_val.reshape(preds.shape[:-1])
if keep_same_device:
stoi_val = stoi_val.to(preds.device)
return stoi_val