hS SSKrSSKJr SSKJrJrJr SSKrSSKJ r SSK J r SSK J r SSKJr SSKJr S rS rS rS /S /S S/S.rS\S\\S\S\\\\44Sjr"SS\ 5rg)N)Path)ListTupleUnion)Tensor)Dataset)download_url_to_file)_get_librispeech_metadata) _extract_tarlibrispeech_finetuningzIhttps://dl.fbaipublicfiles.com/librilight/data/librispeech_finetuning.tgz@5d1efdc777b548194d7e09ba89126e2188026df9fd57aa57eb14408d2b2342afz1h/0z1h/*9h)10min1h10hpathfolders _ext_audioreturnc p[U5n/nUHnURUSU35Vs/sHoURU5PM nnX6Vs/sHAn[URRR5[UR 54PMC sn- nM UR SS9 U$s snfs snf)aGet the file names and the corresponding file paths without `speaker_id` and `chapter_id` directories. The format of path is like: {root}/{_ARCHIVE_NAME}/1h/[0-5]/[clean, other] or {root}/{_ARCHIVE_NAME}/9h/[clean, other] Args: path (Path): Root path to the dataset. folders (List[str]): Folders that contain the desired audio files. _ext_audio (str): Extension of audio files. Returns: List[Tuple[str, str]]: List of tuples where the first element is the relative path to the audio file. The format of relative path is like: 1h/[0-5]/[clean, other] or 9h/[clean, other] The second element is the file name without audio extension. z/*/*/*/*cUSUS-$)Nr)xs ^/var/www/auris/envauris/lib/python3.13/site-packages/torchaudio/datasets/librilight_limited.py$_get_fileids_paths..,s1Q4!A$;)key)rglob relative_tostrparentstemsort)rrr files_pathsfolderppathss r_get_fileids_pathsr*s( :DK.2ii6((:,8W.XY.Xt$.XYEREqQXX__334c!&&kBERR ./ ZRs B. AB3c ~\rSrSrSrSrSrSS\\\ 4S\S\ SS 4S jjr S \ S\ \\ \\ \ \ 44S jrS\ 4S jrSrg )LibriLightLimited0aSubset of Libri-light :cite:`librilight` dataset, which was used in HuBERT :cite:`hsu2021hubert` for supervised fine-tuning. Args: root (str or Path): Path to the directory where the dataset is found or downloaded. subset (str, optional): The subset to use. Options: [``"10min"``, ``"1h"``, ``"10h"``] (Default: ``"10min"``). download (bool, optional): Whether to download the dataset if it is not found at root path. (default: ``False``). z .trans.txtz.flacrootsubsetdownloadrNcU[;a#[S[R5SU35e[Un[R"U5n[R R U[5Ul[R R U[S35n[R RUR5(dTU(d [S5e[R RU5(d[[U[S9 [U5 [!URX@R"5Ulg)Nz`subset` must be one of z . Found: z.tgzz9Dataset not found. Please use `download=True` to download) hash_prefix) _SUBSET_MAP ValueErrorkeysosfspathrjoin _ARCHIVE_NAME_pathisdir RuntimeErrorisfiler _URL _CHECKSUMr r*r_fileids_paths)selfr.r/r0rarchives r__init__LibriLightLimited.__init__?s  $7 8H8H8J7K9U[T\]^ ^f%yyWW\\$ 6 '',,t d%;<ww}}TZZ(("#^__77>>'**$T7 J  !0WooVrncURUup#[X0RX RUR5n[ R "[RRURUS55upVU4USS-$)a8Load 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: Waveform int: Sample rate str: Transcript int: Speaker ID int: Chapter ID int: Utterance ID rrN) r@r r:r_ext_txt torchaudioloadr6rr8)rArE file_pathfileidmetadatawaveform_s r __getitem__LibriLightLimited.__getitem__Tsn*!//2 ,VZZOO]a]j]jk oobggll4::x{&KL {Xab\))rc,[UR5$)N)lenr@)rAs r__len__LibriLightLimited.__len__ns4&&''r)r@r:)rF)__name__ __module__ __qualname____firstlineno____doc__rGrrr"rboolrCintrrrOrS__static_attributes__rrrr,r,0s HJ  WCIWW W  W**S*U63S#s+J%K*4((rr,)r6pathlibrtypingrrrrHtorchrtorch.utils.datartorchaudio._internalr torchaudio.datasets.librispeechr torchaudio.datasets.utilsr r9r>r?r3r"r*r,rrrrds %%$5E2) R N &$H TDI34PUVY[^V^P_K`:?(?(r