+ j5L^RIt^RIHt^RIHt^RIt^RIHtH t H t ^RI H t H t ] !4'd^RIt] !4'd^RIt^RIHtHt^RIHt^RIHtHtHtHt^R IHtHt!R R ]R R 7t!RR]R R 7t] !RR7]!RR]444t R.t!R#)N)Path)Any)auto_docstringis_soundfile_availableis_torch_available)requiresrequires_backends) AudioInputmake_list_of_audio) BatchFeature) AudioKwargsProcessingKwargsProcessorMixinUnpack)PreTokenizedInput TextInputc0a]tRt^%toRtV3RltRtVtR#)CsmAudioKwargsa encoded_length_kwargs (`dict[str, Any]`, *optional*): Dictionary of keyword arguments used to compute the encoded audio sequence length. This includes parameters such as `kernel_sizes`, `strides`, `dilations`, and `use_causal_conv` that define the convolutional layers used in audio encoding. The encoded length is used to determine how many audio tokens to generate for each audio input in the text sequence. cJ<V^8dQh/S[S[S[3,R,;R&#)Nencoded_length_kwargs)dictstrr)format __classdict__s"g/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/csm/processing_csm.py __annotate__CsmAudioKwargs.__annotate__%s S>D00N)__name__ __module__ __qualname____firstlineno____doc____annotate_func____static_attributes____classdictcell__rs@rrr%srrF)totalc pa]tRt^1toRRRRRRR/RR R .ROR .ROR .ROR R/RR/RRR//tV3RltRtVtR#)CsmProcessorKwargs text_kwargspaddingT padding_sideleftadd_special_tokensF audio_kwargsr kernel_sizesstrides dilationsuse_causal_conv sampling_ratei] common_kwargsreturn_tensorsptc&<V^8dQh/S[;R&#)rr2)r)rrs"rrCsmProcessorKwargs.__annotate__1s  rr N)r>r? r>r? r>r?r>)r?r?r?rDr?r?r?r?r?r?r@r?r)r?r?r?r?r?r?r?r?r?r?r?r?r?r?r?)r!r"r#r$ _defaultsr&r'r(r)s@rr,r,1si  t F %  # PHJ!4 & U  *D1Irr,)backendscaa]tRt^Fto]tR V3Rllt]R Rl4tV3RlRlt ] R V3RlRll4t ] R4t RtVtV;t#) CsmProcessorc<\VR4'g)RVnVPVP4VnM"VPVnVPVn\VR4'g)RVnVPVP4VnM"VPVnVP Vn\ SV`WVR7R#) audio_tokenz <|AUDIO|>audio_eos_tokenz <|audio_eos|>) chat_templateN)hasattrrLconvert_tokens_to_idsaudio_token_idrMaudio_eos_token_idsuper__init__)selffeature_extractor tokenizerrN __class__s&&&&rrTCsmProcessor.__init__Ks y-00*D "+"A"A$BRBR"SD (44D "+":":D y"344#2D &/&E&EdFZFZ&[D ##,#<#*CsmProcessor.save_audio..s#@qep`aAPSUY{A[A[eps#%FTzAInvalid input path. Please provide a string, or a list of stringsz5The number of audio and saving paths must be the samer2r7N)r r rvrrrptupleall ValueErrorlen _merge_kwargsr,r[torchTensorcpufloatnumpysfwrite) rUrmrnro output_kwargsr2r7 audio_valuerxs &&&, r save_audioCsmProcessor.save_audios $ .#5) kC; / /&-K[4-88SS@qep@qSSS@qep@q=q=q`a a u:[) )TU U**   %^4 $_5 !%5NK+u||44)oo/557==? HHQ] 36rc <V^8dQhRS[S[,S[S[,,S[S[,,R,RS[R,RS[R,RS[R,RS[S[,/#)rtextNrm output_labelsdepth_decoder_labels_ratioro)rrrpr boolrrr,)rrs"rrrqs~OCOC++d9o=EV@WWZ^^OCD OCd{ OC %*DL OC +, OCrc VP!\3RVPP/VBpVR,pVR,pVP RR4p V R8wd#\ VP P R24h\V\4'dV.pMa\V\\34'd;\;QJdRV4F 'dK R M R M !RV44'g \ R 4hVU u.uFqPVP4NK p p ^p Ve\V4p\!V4p \#V 4^8d2V \#V 48wd"Vf \ R 4h\ R V RV R24hVEeVP%R/4p VUu.uF&qP&!VP(R,3/V BNK( ppVP+4p.pVFp.pVPV9dUVP%^4pVPV,pVP-V4VP/VPR^4pKeRV9d%VP/RVP%^4^4pK+VP-V4K TpVP!V3/VBp/pVP1V4VEeVP%RR4..pp^pV EF@p V ^8XdNVP-\2P4!^44VP-\6P8!R.44KXTP-\2P:!VVVV ,Uu.uFCp\V\6P<4'dVP?4PA4MTNKE upRR74TP-\6P8!VVVV ,Uu.uFpVP(R,NK up4PCRR74VV , pEKC VPD!V3/VBpVP%RR4VP1V4\GRV44pVUu.uFIp\6PHPJPMV^VVP(R,, 3RR7NKK pp\6PN!V^R7VR&V'dVR,VPP8HPS4pVP(^,p VR8:d;\6PT!V 4R\WV ^V, ,4p!VV!,p"MTp"\6PX!VR,VPP8HVR,VPZ8H,VR,R4p#RV#V"R ,V"R!,3&V#VR&\]VV R7#uup iuupiuupiuupiuupi)"a output_labels (bool, *optional*, default=False): Whether to return labels for training. Indices will be in `[config.audio_token_id, -100, -101]`. - `config.audio_token_id` indicates an audio frame (considering sequence length elements as frames) - `-100` will be ignored in the loss computation - `-101` indicates the audio frame will be used only for the backbone model (using the first codebook token as labels) depth_decoder_labels_ratio (float, *optional*, default=1.0): The ratio of audio frames to keep for the depth decoder labels. Returns: [`BatchFeature`]: A [`BatchFeature`] with the following fields: - **input_ids** -- List of token ids to be fed to a model. Returned when `text` is not `None`. - **input_values** -- List of audio values to be fed to a model. Returned when `audio` is not `None`. - **attention_mask** -- List of indices specifying which tokens should be attended to by the model (when `return_attention_mask=True` or if *"attention_mask"* is in `self.model_input_names` and if `text` is not `None`). - **labels** -- List of labels for the audio frames. Returned when `output_labels=True`. tokenizer_init_kwargsr-r2r9Nr:z% only supports `return_tensors='pt'`.c3B"TFp\V\4xK R#5iru)rvr)rwts& rry(CsmProcessor.__call__..s9[VZQR*Q:L:LVZsFTzAInvalid input text. Please provide a string, or a list of stringsz@No audio were provided, but there are audio tokens in the promptz)The number of audio tokens in each text (z7) should be the same as the number of provided audios (z).rz return_attention_mask)axis)dim padding_maskc3F"TFqPR,xK R#5i)r?N)shape)rwcut_idxss& rryrsR=Q..,,=Qs!)valueinput_values_cutoffs input_ids?labels)data tensor_typerii)NNNr)rr?)/rr,rW init_kwargsgetr}rXr!rvrrpr{r|countrLr r~sumpoprjrcopyappendreplaceupdatenpzerosrtensor concatenaterrrcumsumrVmaxnn functionalpadstackrQnonzerorandpermintwhererRr )$rUrrmrrrorr-r2r9rn_audio_in_textn_audior audio_arraynum_audio_tokens_listnum_audio_tokens_list_copy expanded_textsample replace_strnum_audio_tokensexpanded_audio_tokenencodingrconcatenated_audioroffsetel audio_inputsmax_lenraudio_frame_idxsn_audio_frames rand_idxsskip_frames_idxsrs$&&&&&, r__call__CsmProcessor.__call__s}:**  "&.."<"<  $M2 $^4 $)94@ T ! 7 788]^_ _ dC 6DTD%=11cc9[VZ9[ccc9[VZ9[6[6[`a a>BCd774#3#34dC  &u-E%jG  ! #33G(G} !cdd ??PQ229">  $0$4$45Lb$Q !lq%lq]h(():):2)>XBWXlq "%*?)C)C)E &M &&&0'A'E'Ea'H$+/+;+;>N+N(&&';<#^^D,<,>$6+6 H     4d ;792 4 F*a<&--bhhqk:(// bT0BC&--+0'9I*J*JB5?r5<<4P4P 0VX X*J"$ )// U6FU\L\=]%^=]rbhhrll=]%^_ffkmfng%F#+& 112DU UL   ^T 2 KK %R=QRRG!5$ 4H##''1gr@R6R2S[]'^ 4 !$,1;;7KQR+SD' (  $[ 1T5H5H HQQS -33A6N)S0!NN>:;sSSTWqSqAr=st #3I#> #3 [[k"d&9&99d;>OSWSjSj>jk[!F FJF#D)+;D+AA B#DN>BBGD$%L&_$s :#V>,W A W W AWcVPPpVPPpVUu.uF q3R8wgK VNK pp\W,R.,4#uupi)rr)rWmodel_input_namesrVrp)rUtokenizer_input_namesfeature_extractor_input_namesnames& rrCsmProcessor.model_input_names:sd $ @ @(,(>(>(P(P%;X(r:W$cq[q:W%(r)IMcLddee)ss AA)rMrRrLrQru)NNNN)NFr)r!r"r#r$r,valid_processor_kwargsrT staticmethodrjrrrpropertyrr'r( __classcell__)rXrs@@rrJrJFso0T,##J44BOCOCOCbfffrrJ)r)"r\pathlibrtypingrrrutilsrrrutils.import_utilsrr rrsr audio_utilsr r feature_extraction_utilsr processing_utilsr rrrtokenization_utils_baserrrr,rJ__all__r rrrs OO=94UUC 1[ 1)* :zf>zfzfz  r