o yiw@s^ddlmZddlmZmZddlmZddlmZddl m Z ddgiZ GdddeZ d S) )Any)Tensortensor)$perceptual_evaluation_speech_quality)Metric)_PESQ_AVAILABLE!PerceptualEvaluationSpeechQualitypesqc seZdZUdZeed<eed<dZeed<dZeed<dZ eed< dd e d e d e d e ddf fdd Z dededdfddZdefddZZS)ra Calculates `Perceptual Evaluation of Speech Quality`_ (PESQ). It's a recognized industry standard for audio quality that takes into considerations characteristics such as: audio sharpness, call volume, background noise, clipping, audio interference ect. PESQ returns a score between -0.5 and 4.5 with the higher scores indicating a better quality. This metric is a wrapper for the `pesq package`_. Note that input will be moved to ``cpu`` to perform the metric calculation. As input to ``forward`` and ``update`` the metric accepts the following input - ``preds`` (:class:`~torch.Tensor`): float tensor with shape ``(...,time)`` - ``target`` (:class:`~torch.Tensor`): float tensor with shape ``(...,time)`` As output of `forward` and `compute` the metric returns the following output - ``pesq`` (:class:`~torch.Tensor`): float tensor with shape ``(...,)`` of PESQ value per sample .. note:: using this metrics requires you to have ``pesq`` install. Either install as ``pip install torchmetrics[audio]`` or ``pip install pesq``. ``pesq`` will compile with your currently installed version of numpy, meaning that if you upgrade numpy at some point in the future you will most likely have to reinstall ``pesq``. Args: fs: sampling frequency, should be 16000 or 8000 (Hz) mode: ``'wb'`` (wide-band) or ``'nb'`` (narrow-band) keep_same_device: whether to move the pesq value to the device of preds n_processes: integer specifiying the number of processes to run in parallel for the metric calculation. Only applies to batches of data and if ``multiprocessing`` package is installed. kwargs: Additional keyword arguments, see :ref:`Metric kwargs` for more info. Raises: ModuleNotFoundError: If ``pesq`` package is not installed ValueError: If ``fs`` is not either ``8000`` or ``16000`` ValueError: If ``mode`` is not either ``"wb"`` or ``"nb"`` Example: >>> from torchmetrics.audio.pesq import PerceptualEvaluationSpeechQuality >>> import torch >>> g = torch.manual_seed(1) >>> preds = torch.randn(8000) >>> target = torch.randn(8000) >>> nb_pesq = PerceptualEvaluationSpeechQuality(8000, 'nb') >>> nb_pesq(preds, target) tensor(2.2076) >>> wb_pesq = PerceptualEvaluationSpeechQuality(16000, 'wb') >>> wb_pesq(preds, target) tensor(1.7359) sum_pesqtotalFfull_state_updateis_differentiableThigher_is_betterfsmode n_processeskwargsreturnNc stjd i|tstd|dvrtd|||_|dvr(td|||_t|ts;|dkr;td|||_ |j dt d d d |j d t dd d dS)NzPerceptualEvaluationSpeechQuality metric requires that `pesq` is installed. Either install as `pip install torchmetrics[audio]` or `pip install pesq`.)i@i>z:Expected argument `fs` to either be 8000 or 16000 but got )wbnbz;Expected argument `mode` to either be 'wb' or 'nb' but got rzCExpected argument `n_processes` to be an int larger than 0 but got r gsum)defaultdist_reduce_fxr ) super__init__rModuleNotFoundError ValueErrorrr isinstanceintr add_stater)selfrrrr __class__rK/home/ubuntu/.local/lib/python3.10/site-packages/torchmetrics/audio/pesq.pyrTs z*PerceptualEvaluationSpeechQuality.__init__predstargetcCsJt|||j|jd|j|jj}|j|7_|j| 7_dS)z*Update state with predictions and targets.FN) rrrrtor devicerr numel)r"r&r' pesq_batchrrr%updatens z(PerceptualEvaluationSpeechQuality.updatecCs |j|jS)zComputes metric.)r r )r"rrr%computews z)PerceptualEvaluationSpeechQuality.compute)r)__name__ __module__ __qualname____doc__r__annotations__r boolr rr strrrr,r- __classcell__rrr#r%rs* 4    N) typingrtorchrr"torchmetrics.functional.audio.pesqrtorchmetrics.metricrtorchmetrics.utilities.importsr__doctest_requires__rrrrr%s