o iK @sddlmZddlmZmZmZddlmZddlm Z m Z er6ddl Z ddl mZddlmZddlmZGd d d ZGd d d ed dZGddded dZ  d"d#ddZed krddlZddlZeZed!eZej ej!Z"ede"Z#e$e#dSdS)$) annotations) TYPE_CHECKINGLiteral TypedDict) AudioData)TranscribeOutputBaseWhisperCompatibleRecognizerN WhisperModel)Segment)Unpackc@s eZdZd ddZdd d Zd S)TranscribableAdaptermodelr returnNonecCs ||_dSN)r)selfrrn/home/kuhnn/.local/lib/python3.10/site-packages/speech_recognition/recognizers/whisper_local/faster_whisper.py__init__s zTranscribableAdapter.__init__ audio_array np.ndarrayTranscribeOutputBase[Segment]cKs>|jj|fi|\}}t|}ddd|D||jdS)N css|]}|jVqdSr)text).0segmentrrr sz2TranscribableAdapter.transcribe..)rsegmentslanguage)r transcribelistjoinr)rrkwargssegments_generatorinforrrrr s zTranscribableAdapter.transcribeN)rr rr)rrrr)__name__ __module__ __qualname__rr rrrrr s r c@s&eZdZUded<ded<ded<dS)InitOptionalParameterszLiteral['cpu', 'gpu', 'auto']devicestr compute_type download_rootNr&r'r(__annotations__rrrrr)!  r)F)totalc@s&eZdZUded<ded<ded<dS)TranscribeOptionalParametersr+rz"Literal['transcribe', 'translate']taskint beam_sizeNr.rrrrr2)r0r2base audio_datarrr+ show_dictbool init_optionsInitOptionalParameters | Nonetranscribe_options$Unpack[TranscribeOptionalParameters]r#str | TranscribeOutputBase[Segment]cKsBddlm}||fi|p i}tt|}|j|fd|i|S)aPerforms speech recognition on ``audio_data`` (an ``AudioData`` instance), using Whisper. Pick ``model`` size (Same as Whisper). If ``show_dict`` is true, returns the detailed response from Whisper, including the detected language. Otherwise returns only the transcription. You can specify: * ``language``: recognition language, an uncapitalized 2 letters language name like "en" or "fr". * If not set, Faster Whisper will automatically detect the language. * ``task`` * If you want transcribe + **translate** to english, set ``task="translate"``. Other values are passed directly to whisper. See https://github.com/SYSTRAN/faster-whisper/blob/master/faster_whisper/transcribe.py for all options. rr r8)faster_whisperr rr recognize) recognizerr7rr8r:r<r whisper_recognizerrrrr@1s r@__main__ audio_file)r6FN) r7rrr+r8r9r:r;r<r=rr>)% __future__rtypingrrrspeech_recognition.audior1speech_recognition.recognizers.whisper_local.baserrnumpynpr?r faster_whisper.transcriber typing_extensionsr r r)r2r@r&argparsespeech_recognitionsrArgumentParserparser add_argument parse_argsargs from_filerDr7 transcriptionprintrrrrs4       %