o 헦if@sddlZddlZddlmZddlmZmZmZmZddl m Z ddl m Z dddZ hd Zd eed ed eefd dZGddde ZdS)N)Path)IterableListTupleUnion)Dataset)download_url_to_file@209a8b4cd265013e96f4658632a9878103b0c5abf62b50d4ef3ae1be226b29e4@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>8.DOT+PLUS-DASH"QUOTE'QUOTE(PAREN)PAREN,COMMA--DASH.POINT/SLASH:COLON{BRACE(PARENS)PARENS-HYPHEN.PERIOD"UNQUOTE%PERCENT.DECIMAL "END-QUOTE "IN-QUOTES #HASH-MARK &ERSAND 'END-QUOTE )END-PAREN .FULL-STOP #POUND-SIGN #SHARP-SIGN (LEFT-PAREN )END-PARENS ...ELLIPSIS ;SEMI-COLON {LEFT-BRACE {OPEN-BRACE "CLOSE-QUOTE 'INNER-QUOTE (PARENTHESES )CLOSE-PAREN )RIGHT-PAREN }CLOSE-BRACE }RIGHT-BRACE "DOUBLE-QUOTE "END-OF-QUOTE 'SINGLE-QUOTE (BEGIN-PARENS)END-THE-PAREN;SEMI-COLON(1)?QUESTION-MARK(IN-PARENTHESES)UN-PARENTHESES'END-INNER-QUOTE)END-PARENTHESES(OPEN-PARENTHESES!EXCLAMATION-POINT)CLOSE-PARENTHESESlinesexclude_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)rErF_alt_recmudictlinewordphonesrWY/home/ubuntu/transcripts/venv/lib/python3.10/site-packages/torchaudio/datasets/cmudict.py_parse_dictionaryJs$    rYc@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_symbolsrootrFr[r\r]rGNc 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|]}|qSrW)rM).0rTrWrWrX sz$CMUDict.__init__..zlatin-1)encoding)rF)rFr _root_pathospathisdir RuntimeErrorbasenameexists _CHECKSUMSgetropen readlines_symbolsrY _dictionary) selfr^rFr[r\r] dict_file symbol_filechecksumtextrWrWrX__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 )ro)rprvrWrWrX __getitem__s zCMUDict.__getitem__cCs t|jS)N)lenrorprWrWrX__len__s zCMUDict.__len__cCs |jS)zLlist[str]: A list of phonemes symbols, such as ``"AA"``, ``"AE"``, ``"AH"``.)rncopyryrWrWrXsymbolss zCMUDict.symbols)T)__name__ __module__ __qualname____doc__rstrrboolruintrrrwrzpropertyr|rWrWrWrXrZis0  )rZ)rdrJpathlibrtypingrrrrtorch.utils.datartorchaudio._internalrrjrOrrrYrZrWrWrWrXs   <