o *if@sddlZddlZddlmZddlmZmZmZmZddl m Z ddl m Z dddZ hd Zd eed ed eefd dZGddde ZdS)N)Path)IterableListTupleUnion)Dataset)download_url_to_fileZ@209a8b4cd265013e96f4658632a9878103b0c5abf62b50d4ef3ae1be226b29e4Z@408ccaae803641c6d7b626b6299949320c2dbca96b2220fd3fb17887b023b027)Bhttp://svn.code.sf.net/p/cmusphinx/code/trunk/cmudict/cmudict-0.7bJhttp://svn.code.sf.net/p/cmusphinx/code/trunk/cmudict/cmudict-0.7b.symbols>8z.DOTz"UNQUOTEz )END-PARENSz #POUND-SIGNz)PARENSz)PARENz }RIGHT-BRACEz,COMMAz)END-THE-PARENz)UN-PARENTHESESz.DECIMALz;SEMI-COLON(1)z "END-QUOTEz)CLOSE-PARENTHESESz/SLASHz (LEFT-PARENz(OPEN-PARENTHESESz--DASHz (BEGIN-PARENSz )END-PARENz (PARENTHESESz )RIGHT-PARENz "END-OF-QUOTEz"QUOTEz &ERSANDz?QUESTION-MARKz ;SEMI-COLONz(PARENz "DOUBLE-QUOTEz #SHARP-SIGNz+PLUSz "CLOSE-QUOTEz 'INNER-QUOTEz'END-INNER-QUOTEz%PERCENTz{BRACEz .FULL-STOPz {LEFT-BRACEz "IN-QUOTESz {OPEN-BRACEz.PERIODz!EXCLAMATION-POINTz ...ELLIPSISz )CLOSE-PARENz 'END-QUOTEz #HASH-MARKz 'SINGLE-QUOTEz(PARENSz)END-PARENTHESESz-DASHz.POINTz(IN-PARENTHESESz-HYPHENz'QUOTEz }CLOSE-BRACEz:COLONlinesexclude_punctuationsreturncCstd}g}|D]A}|r|drq |d\}}|tvr7|r#q |dr+d}n |dr3d}n|d}t|d|}|d}|||fq |S) Nz \([0-9]+\)z;;;z z...z--r )recompile startswithstripsplit _PUNCTUATIONSsubappend)r r Z_alt_reZcmudictlinewordZphonesrj/var/www/html/eduruby.in/lip-sync/lip-sync-env/lib/python3.10/site-packages/torchaudio/datasets/cmudict.py_parse_dictionaryJs$    rc@seZdZdZ ddddddeeefded ed ed ed d f ddZde d e ee effddZ d e fddZ ed e efddZd S)CMUDictaZ*CMU Pronouncing Dictionary* :cite:`cmudict` (CMUDict) dataset. Args: root (str or Path): Path to the directory where the dataset is found or downloaded. exclude_punctuations (bool, optional): When enabled, exclude the pronounciation of punctuations, such as `!EXCLAMATION-POINT` and `#HASH-MARK`. download (bool, optional): Whether to download the dataset if it is not found at root path. (default: ``False``). url (str, optional): The URL to download the dictionary from. (default: ``"http://svn.code.sf.net/p/cmusphinx/code/trunk/cmudict/cmudict-0.7b"``) url_symbols (str, optional): The URL to download the list of symbols from. (default: ``"http://svn.code.sf.net/p/cmusphinx/code/trunk/cmudict/cmudict-0.7b.symbols"``) TFr r )downloadurl url_symbolsrootr rrr r Nc CsB||_t||_tj|jstd||jtj|}|jtj|}tj|sC|s7td|t |d}t |||tj|s^|sRtd|t |d}t |||t |d} dd| D|_Wdn1sxwYt |ddd} t| |jd |_WddS1swYdS) Nz#The root directory does not exist; z`The dictionary file is not found in the following location. Set `download=True` to download it. z\The symbol file is not found in the following location. Set `download=True` to download it. rcSsg|]}|qSr)r).0rrrr sz$CMUDict.__init__..zlatin-1)encoding)r )r rZ _root_pathospathisdir RuntimeErrorbasenameexists _CHECKSUMSgetropen readlines_symbolsr _dictionary) selfr!r rrr Z dict_fileZ symbol_fileZchecksumtextrrr__init__{s<        "zCMUDict.__init__ncCs |j|S)aLoad the n-th sample from the dataset. Args: n (int): The index of the sample to be loaded. Returns: Tuple of a word and its phonemes str: Word List[str]: Phonemes )r1)r2r5rrr __getitem__s zCMUDict.__getitem__cCs t|jS)N)lenr1r2rrr__len__s zCMUDict.__len__cCs |jS)zLlist[str]: A list of phonemes symbols, such as ``"AA"``, ``"AE"``, ``"AH"``.)r0copyr8rrrsymbolss zCMUDict.symbols)T)__name__ __module__ __qualname____doc__rstrrboolr4intrrr6r9propertyr;rrrrris0  )r)r&rpathlibrtypingrrrrZtorch.utils.datarZtorchaudio._internalrr,rr@rArrrrrrs   <