a hf@sddlZddlZddlmZddlmZmZmZmZddl m Z ddl m Z dddZ hd Zeeeeed d d ZGd dde 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 )CLOSE-PARENz "CLOSE-QUOTEz'END-INNER-QUOTEz )RIGHT-PARENz (LEFT-PARENz{BRACEz (BEGIN-PARENSz-HYPHENz 'INNER-QUOTEz (PARENTHESESz.PERIODz }RIGHT-BRACEz #HASH-MARKz%PERCENTz:COLONz'QUOTEz {LEFT-BRACEz ;SEMI-COLONz(PARENSz "END-OF-QUOTEz+PLUSz "END-QUOTEz(PARENz)END-PARENTHESESz--DASHz)UN-PARENTHESESz.POINTz )END-PARENz 'SINGLE-QUOTEz )END-PARENSz;SEMI-COLON(1)z "DOUBLE-QUOTEz.DOTz!EXCLAMATION-POINTz &ERSANDz(OPEN-PARENTHESESz-DASHz }CLOSE-BRACEz 'END-QUOTEz)CLOSE-PARENTHESESz ...ELLIPSISz"QUOTEz)PARENz.DECIMALz)END-THE-PARENz {OPEN-BRACEz #SHARP-SIGNz .FULL-STOPz #POUND-SIGNz/SLASHz)PARENSz(IN-PARENTHESESz"UNQUOTEz "IN-QUOTESz?QUESTION-MARKz,COMMA)linesexclude_punctuationsreturncCstd}g}|D]}|r|dr&q|d\}}|tvrn|rFq|drVd}n|drfd}n|d}t|d|}|d}|||fq|S) Nz \([0-9]+\)z;;;z z...z--r )recompile startswithstripsplit _PUNCTUATIONSsubappend)r r Z_alt_reZcmudictlinewordZphonesrI/var/www/auris/lib/python3.9/site-packages/torchaudio/datasets/cmudict.py_parse_dictionaryJs$    rc @s|eZdZdZdddddeeefeeeeddd d Ze e ee efd d d Z e dddZ ee edddZdS)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_symbolsN)rootr rrr r c CsB||_t||_tj|js,td||jtj|}|jtj|}tj|s|sntd|t |d}t |||tj|s|std|t |d}t |||t |d$} dd| D|_Wdn1s0Yt |ddd$} t| |jd |_Wdn1s40YdS) 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 z$CMUDict.__init__..zlatin-1)encoding)r )r rZ _root_pathospathisdir RuntimeErrorbasenameexists _CHECKSUMSgetropen readlines_symbolsr _dictionary) selfr!r rrr Z dict_fileZ symbol_fileZchecksumtextrrr__init__{s8         2zCMUDict.__init__)nr cCs |j|S)aLoad the n-th sample from the dataset. Args: n (int): The index of the sample to be loaded. Returns: Tuple of a word and its phonemes str: Word List[str]: Phonemes )r2)r3r6rrr __getitem__szCMUDict.__getitem__)r cCs t|jS)N)lenr2r3rrr__len__szCMUDict.__len__cCs |jS)zLlist[str]: A list of phonemes symbols, such as ``"AA"``, ``"AE"``, ``"AH"``.)r1copyr9rrrsymbolsszCMUDict.symbols)T)__name__ __module__ __qualname____doc__rstrrboolr5intrrr7r:propertyr<rrrrris$  )r)r'rpathlibrtypingrrrrZtorch.utils.datarZtorchaudio._internalrr-rrArBrrrrrrs   <