+
    œ°³jJ  ã                   ó²   € ^ RI Ht ^RIHt ^RIHt ^RIHtHt ]! RR7      ] ! R R]4      4       4       t	]! RR7      ] ! R	 R
]4      4       4       t
R
R.tR# )é    )Ústrict)ÚPreTrainedConfig)Úauto_docstring)ÚCONFIG_MAPPINGÚ
AutoConfigzvibevoice/VibeVoice-1.5B-hf)Ú
checkpointc                   óT   a € ] tR t^t o RtRt^@t^tRtRt	Rt
RtRtRtV 3R	 ltR
tV tR# )ÚVibeVoiceDiffusionHeadConfiga¡  
latent_size (`int`, *optional*, defaults to 64):
    Dimensionality of the acoustic latents the head denoises.
frequency_embedding_size (`int`, *optional*, defaults to 256):
    The size of the sinusoidal frequency embedding for timestep encoding in the diffusion head.
diffusion_max_period (`int`, *optional*, defaults to 10000):
    The maximum period for the sinusoidal frequency embedding in the diffusion head.
i   i   gñhãˆµøä>Úsilué   i'  Fc                ó†   <€ V ^8„  d   Qh/ S[ ;R&   S[ ;R&   S[ ;R&   S[ ;R&   S[;R&   S[;R&   S[ ;R&   S[ ;R&   S[;R	&   # )
é   Úhidden_sizeÚlatent_sizeÚnum_hidden_layersÚintermediate_sizeÚrms_norm_epsÚ
hidden_actÚfrequency_embedding_sizeÚdiffusion_max_periodÚmlp_bias)ÚintÚfloatÚstrÚbool)ÚformatÚ__classdict__s   "€Úv/home/luca/screen-checking/.venv/lib/python3.14/site-packages/transformers/models/vibevoice/configuration_vibevoice.pyÚ__annotate__Ú)VibeVoiceDiffusionHeadConfig.__annotate__   sz   ø‡ ‚ ñ Ññ ñ Ññ ñ Ññ ñ Ñ!ñ ñ  Ññ! ñ" Ññ# ñ$ "Ñ'ñ% ñ& Ñ%ñ' ñ( Ñò) ó    © N)Ú__name__Ú
__module__Ú__qualname__Ú__firstlineno__Ú__doc__r   r   r   r   r   r   r   r   r   Ú__annotate_func__Ú__static_attributes__Ú__classdictcell__)r   s   @r   r
   r
      sD   ø‡ € ñð €KØ€KØÐØ!ÐØ€LØ€JØ$'ÐØ %ÐØ€H÷) ƒ r!   r
   c                   óˆ   a a€ ] tR t^-t oRtRtR]R]R]R]/tRt	Rt
RtRtRtRtR	tR
tRtRtV 3R ltR tV3R ltRtVtV ;t# )ÚVibeVoiceConfiga«  
semantic_model_config (`Union[AutoConfig, dict]`, *optional*):
    The config object or dictionary of the semantic tokenizer encoder. This tokenizer extracts semantic features from audio.
diffusion_head_config (`Union[VibeVoiceDiffusionHeadConfig, dict]`, *optional*):
    The config object or dictionary of the diffusion head used to synthesize acoustic latents.
audio_bos_token_id (`int`, *optional*, defaults to 151652):
    The token ID indicating the start of audio tokens.
audio_eos_token_id (`int`, *optional*, defaults to 151653):
    The token ID indicating the end of audio tokens.
diffusion_loss_weight (`float`, *optional*, defaults to 0.5):
    The weight of the diffusion loss in the overall loss computation. The cross entropy loss for the language
    modeling head is weighted by `(1 - diffusion_loss_weight)`.

```python
>>> from transformers import VibeVoiceForConditionalGeneration, VibeVoiceConfig

>>> # Initializing a VibeVoice configuration
>>> configuration = VibeVoiceConfig()

>>> # Initializing a 1.5B model with random weights
>>> model = VibeVoiceForConditionalGeneration(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config
```Ú	vibevoiceÚaudio_configÚsemantic_model_configÚtext_configÚdiffusion_head_configNi[P idP ieP ifP g      à?c                ól  <€ \        V P                  \        4      '       d^   V P                  P                  R R4      V P                  R &   \        V P                  R ,          ,          ! R	/ V P                  B V n        M%V P                  f   \        R,          ! 4       V n        \        V P
                  \        4      '       d^   V P
                  P                  R R4      V P
                  R &   \        V P
                  R ,          ,          ! R	/ V P
                  B V n        M'V P
                  f   \        R,          ! ^€R7      V n        \        V P                  \        4      '       d^   V P                  P                  R R4      V P                  R &   \        V P                  R ,          ,          ! R	/ V P                  B V n        M%V P                  f   \        R,          ! 4       V n        \        V P                  \        4      '       d   \        R	/ V P                  B V n        MHV P                  f;   \        V P                  P                  V P                  P                  R7      V n        V P                  P                  V n
        \        V P                  RR4      V n        \        SV `8  ! R	/ VB  R# )
Ú
model_typeÚvibevoice_acoustic_tokenizerNÚ$vibevoice_acoustic_tokenizer_encoder)r   Úqwen2)r   r   Útie_word_embeddingsFr"   )Ú
isinstancer.   ÚdictÚgetr   r/   r0   r1   r
   r   Ú
vocab_sizeÚgetattrr7   ÚsuperÚ__post_init__)ÚselfÚkwargsÚ	__class__s   &,€r   r>   ÚVibeVoiceConfig.__post_init__]   s  ø€ Ü�d×'Ñ'¬×.Ò.Ø.2×.?Ñ.?×.CÑ.CÀLÐRpÓ.qˆD×Ñ˜lÑ+Ü .¨t×/@Ñ/@ÀÕ/NÖ OÑ dÐRV×RcÑRcÑ dˆDÕØ×ÑÒ&Ü .Ð/MÖ NÓ PˆDÔä�d×0Ñ0´$×7Ò7Ø7;×7QÑ7Q×7UÑ7UØÐDó8ˆD×&Ñ& |Ñ4ô *8¸×8RÑ8RÐS_Õ8`Ö)añ *Ø×,Ñ,ñ*ˆDÕ&ð ×'Ñ'Ò/Ü)7Ð8^Ö)_ÐloÔ)pˆDÔ&ä�d×&Ñ&¬×-Ò-Ø-1×-=Ñ-=×-AÑ-AÀ,ÐPWÓ-XˆD×Ñ˜\Ñ*Ü-¨d×.>Ñ.>¸|Õ.LÖMÑaÐPT×P`ÑP`ÑaˆDÕØ×ÑÒ%Ü-¨gÖ6Ó8ˆDÔä�d×0Ñ0´$×7Ò7Ü)EÑ)cÈ×HbÑHbÑ)cˆDÕ&Ø×'Ñ'Ò/Ü)EØ ×,Ñ,×8Ñ8Àd×FWÑFW×FcÑFcô*ˆDÔ&ð ×*Ñ*×5Ñ5ˆŒÜ#*¨4×+;Ñ+;Ð=RÐTYÓ#ZˆÔ Ü‰ÒÑ' Ô'r!   c                óª  € V P                   P                  V P                  P                  8w  d;   \        RV P                   P                   RV P                  P                   R24      hV P                   P                  V P
                  P                  8w  d;   \        RV P                   P                   RV P
                  P                   R24      hR# )zOPart of `@strict`-powered validation. Validates the architecture of the config.z%`diffusion_head_config.hidden_size` (z() must match `text_config.hidden_size` (z).z%`diffusion_head_config.latent_size` (z)) must match `audio_config.hidden_size` (N)r1   r   r0   Ú
ValueErrorr   r.   )r?   s   &r   Úvalidate_architectureÚ%VibeVoiceConfig.validate_architecture   sÌ   € à×%Ñ%×1Ñ1°T×5EÑ5E×5QÑ5QÔQÜØ7¸×8RÑ8R×8^Ñ8^Ð7_ð `.Ø.2×.>Ñ.>×.JÑ.JÐ-KÈ2ðOóð ð ×%Ñ%×1Ñ1°T×5FÑ5F×5RÑ5RÔRÜØ7¸×8RÑ8R×8^Ñ8^Ð7_ð `/Ø/3×/@Ñ/@×/LÑ/LÐ.MÈRðQóð ñ Sr!   c                ó
  <€ V ^8„  d   Qh/ S[ S[,          R,          ;R&   S[ S[,          R,          ;R&   S[ S[,          R,          ;R&   S[ S[,          R,          ;R&   S[;R&   S[;R&   S[;R&   S[;R	&   S[;R
&   S[;R&   # )r   Nr.   r/   r0   r1   Úpad_token_idÚeos_token_idÚaudio_bos_token_idÚaudio_eos_token_idÚaudio_token_idÚdiffusion_loss_weight)r9   r   r   r   )r   r   s   "€r   r   ÚVibeVoiceConfig.__annotate__-   sÇ   ø‡ ‚ ñJ Ñ)Õ)¨DÕ0Ñ7ñK ñL  Ñ"2Õ2°TÕ9Ñ@ñM ñN Ñ(Õ(¨4Õ/Ñ6ñO ñP  Ñ"2Õ2°TÕ9Ñ@ñQ ñR ÑñS ñT ÑñU ñV Ñ$ñW ñX Ñ$ñY ñZ Ñ ñ[ ñ\ !Ñ&ò] r!   )r.   r1   r/   r0   r7   r;   )r#   r$   r%   r&   r'   r3   r   r
   Úsub_configsr.   r/   r0   r1   rH   rI   rJ   rK   rL   rM   r>   rE   r(   r)   r*   Ú__classcell__)rA   r   s   @@r   r,   r,   -   sy   ù‡ € ñð4 €Jà˜
Ø Ø�zØÐ!=ð	€Kð 48€LØ<@ÐØ26€KØ<@ÐØ€LØ€LØ$ÐØ$ÐØ €NØ#&Ðõ (òD÷e … r!   r,   N)Úhuggingface_hub.dataclassesr   Úconfiguration_utilsr   Úutilsr   Úautor   r   r
   r,   Ú__all__r"   r!   r   Ú<module>rV      sx   ðõ /å 3Ý #ß -ñ Ð8Ô9ØôÐ#3ó ó ó :ðñ* Ð8Ô9Øô[Ð&ó [ó ó :ð[ð| Ð<Ð
=‚r!   