o i@s|ddlmZddlmZmZmZddlZddlmZddl m Z ddl m Z ddl mZdZd Zd d hZGd d d e ZdS))Path)DictTupleUnionN)Tensor)Dataset)download_url_to_file) _extract_zipzBhttps://datashare.ed.ac.uk/bitstream/handle/10283/3038/DR-VCTK.zip@781f12f4406ed36ed27ae3bce55da47ba176e2d8bae67319e389e07b2c9bd769traintestc @seZdZdZ ddeddeeefdededed d f d d Z d ed e ee ee fffddZ ded e ee ee eeee ffddZde d e ee ee eeee ffddZd e fddZd S)DR_VCTKa?*Device Recorded VCTK (Small subset version)* :cite:`Sarfjoo2018DeviceRV` dataset. Args: root (str or Path): Root directory where the dataset's top level directory is found. subset (str): The subset to use. Can be one of ``"train"`` and ``"test"``. (default: ``"train"``). download (bool): Whether to download the dataset if it is not found at root path. (default: ``False``). url (str): The URL to download the dataset from. (default: ``"https://datashare.ed.ac.uk/bitstream/handle/10283/3038/DR-VCTK.zip"``) r F)downloadurlrootsubsetrrreturnNcCs|tvrtd|dtt|}|d}||_|dd|_|jd|jd|_|jd|jd|_|jd|jd |_|j s_| sZ|sStd t ||t d t ||||j|_t|j|_dS) Nz The subset 'z/' does not match any of the supported subsets: z DR-VCTK.zipzDR-VCTKclean_ set_wav_16kzdevice-recorded_configurationsz _ch_log.txtz=Dataset not found. Please use `download=True` to download it.) hash_prefix)_SUPPORTED_SUBSETS RuntimeErrorr expanduser_subset_path_clean_audio_dir_noisy_audio_dir_config_filepathis_diris_filer _CHECKSUMr _load_config_configsorted_filename_list)selfrrrrarchiver(Q/home/ubuntu/vllm_env/lib/python3.10/site-packages/torchaudio/datasets/dr_vctk.py__init__s&   zDR_VCTK.__init__filepathc Cs|jdkrdnd}i}t|+}t|D]\}}||ks|sq|d\}}} |t| f||<qWd|S1s=wY|S)Nr  )ropen enumeratestripsplitint) r&r+ skip_rowsconfigfilinefilenamesource channel_idr(r(r)r"<s   zDR_VCTK._load_configr9c Csj|ddd\}}|j|\}}|j|}|j|}t|\}} t|\} } || | | ||||fS)N.r_)r2r#rr torchaudioload) r&r9 speaker_id utterance_idr:r;file_clean_audiofile_noisy_audiowaveform_cleansample_rate_cleanwaveform_noisysample_rate_noisyr(r(r)_load_dr_vctk_itemIs  zDR_VCTK._load_dr_vctk_itemncCs|j|}||S)aOLoad the n-th sample from the dataset. Args: n (int): The index of the sample to be loaded Returns: Tuple of the following items; Tensor: Clean waveform int: Sample rate of the clean waveform Tensor: Noisy waveform int: Sample rate of the noisy waveform str: Speaker ID str: Utterance ID str: Source int: Channel ID )r%rH)r&rIr9r(r(r) __getitem__[s  zDR_VCTK.__getitem__cCs t|jS)N)lenr%)r&r(r(r)__len__xs zDR_VCTK.__len__)r )__name__ __module__ __qualname____doc___URLrstrrboolr*rrr3r"rrHrJrLr(r(r(r)r s*  " & &r )pathlibrtypingrrrr>torchrtorch.utils.datartorchaudio._internalrtorchaudio.datasets.utilsr rQr!rr r(r(r(r)s