o
    iz                  	   @   s   d dl Z d dlZd dlmZ d dlZd dlZd dlmZ d dlm	Z	 d dl
mZ d dlmZmZmZmZmZmZmZ d dlmZ G dd	 d	Zed
krye Zejeeddddeeddeedddd\ZZZedej  dS dS )    N)files)cached_path)	get_class)	OmegaConf)infer_process
load_modelload_vocoderpreprocess_ref_audio_text remove_silence_for_generated_wavsave_spectrogram
transcribe)seed_everythingc                   @   sf   e Zd Z								dddZddd	ZdddZdd Zeedddddddd
dddfddZ	dS )F5TTSF5TTS_v1_Base eulerTNc	              
   C   sX  t ttdd| d}	td|	jj }
|	jj}|	jj	j
| _
|	jj	j| _|| _|| _|d ur7|| _ndd l}|j rBdn|j rIdn	|jj rQdnd	| _t| j
|d u|| j|| _d
\}}}|dkr|| j
dkrrd}n| j
dkr{d}d}n|dkrd}d}|sttd| d| d| d| |d}t|
||| j
|| j| j| j| _d S )Nf5_ttszconfigs/z.yamlzf5_tts.model.r   cudaxpumpscpu)zF5-TTSi safetensors
F5TTS_BasevocosiO bigvganF5TTS_Base_bigvganpt
E2TTS_BasezE2-TTSzhf://SWivid//z/model_.)	cache_dir)r   loadstrr   joinpathr   modelbackbonearchmel_specmel_spec_typetarget_sample_rate
ode_methoduse_emadevicetorchr   is_availabler   backendsr   r   vocoderr   r   	ema_model)selfr$   	ckpt_file
vocab_filer*   r+   vocoder_local_pathr,   hf_cache_dir	model_cfg	model_cls	model_arcr-   	repo_name	ckpt_step	ckpt_type r=   >/home/ubuntu/.local/lib/python3.10/site-packages/f5_tts/api.py__init__   sN    



"
zF5TTS.__init__c                 C   s
   t ||S N)r   )r2   	ref_audiolanguager=   r=   r>   r   V   s   
zF5TTS.transcribeFc                 C   s$   t ||| j |rt| d S d S r@   )sfwriter)   r
   )r2   wav	file_waveremove_silencer=   r=   r>   
export_wavY   s   zF5TTS.export_wavc                 C   s   t || d S r@   )r   )r2   spec	file_specr=   r=   r>   export_spectrogram_   s   zF5TTS.export_spectrogramg?g333333?       g      ?c                 C   s   |d u rt dtj}t| || _t|||d\}}t|||| j| j	| j
|||||
|	|||| jd\}}}|d urA| ||| |d urK| || |||fS )Nr   )	show_info)
rO   progress
target_rmscross_fade_durationnfe_stepcfg_strengthsway_sampling_coefspeedfix_durationr,   )randomrandintsysmaxsizer   seedr	   r   r1   r0   r(   r,   rH   rK   )r2   ref_fileref_textgen_textrO   rP   rQ   rR   rU   rT   rS   rV   rW   rG   rF   rJ   r\   rE   srrI   r=   r=   r>   inferb   s8   
zF5TTS.infer)r   r   r   r   TNNNr@   )F)
__name__
__module____qualname__r?   r   rH   rK   printtqdmra   r=   r=   r=   r>   r      s6    

>
r   __main__r   z%infer/examples/basic/basic_ref_en.wavz2Some call me nature, others call me mother nature.zI don't really care what you call me. I've been a silent spectator, watching species evolve, empires rise and fall. But always remember, I am mighty and enduring.z../../tests/api_out.wavz../../tests/api_out.png)r]   r^   r_   rF   rJ   r\   zseed :)!rX   rZ   importlib.resourcesr   	soundfilerC   rf   r   hydra.utilsr   	omegaconfr   f5_tts.infer.utils_inferr   r   r   r	   r
   r   r   f5_tts.model.utilsr   r   rb   f5ttsra   r"   r#   rE   r`   rI   re   r\   r=   r=   r=   r>   <module>   s0    $	 	