app.py

import gradio as gr
#import requests

from PIL import Image
import os

from share_btn import community_icon_html, loading_icon_html, share_js

token = os.environ.get('HF_TOKEN')
whisper_to_gpt = gr.Blocks.load(name="spaces/fffiloni/whisper-to-chatGPT")
tts = gr.Interface.load(name="spaces/Flux9665/IMS-Toucan")
talking_face = gr.Blocks.load(name="spaces/fffiloni/one-shot-talking-face", api_key=token)

def infer(audio):
    whisper_to_gpt_response = whisper_to_gpt(audio, "translate", fn_index=0)
    #print(gpt_response)
    audio_response = tts(whisper_to_gpt_response[1], "English Text", "English Accent", "English Speaker's Voice", fn_index=0)
    #image = Image.open(r"wise_woman_portrait.png")
    portrait_link = talking_face("wise_woman_portrait.png", audio_response, fn_index=0)
    
    #portrait_response = requests.get(portrait_link, headers={'Authorization': 'Bearer ' + token})
    #print(portrait_response.text)
    
    return whisper_to_gpt_response[0], portrait_link, gr.update(visible=True), gr.update(visible=True), gr.update(visible=True)

title = """
    <div style="text-align: center; max-width: 500px; margin: 0 auto;">
        <div
        style="
            display: inline-flex;
            align-items: center;
            gap: 0.8rem;
            font-size: 1.75rem;
            margin-bottom: 10px;
        "
        >
        <h1 style="font-weight: 600; margin-bottom: 7px;">
            GPT Talking Portrait
        </h1>
        </div>
        <p style="margin-bottom: 10px;font-size: 94%;font-weight: 100;line-height: 1.5em;">
        Use Whisper to ask, alive portrait responds  !
        </p>
    </div>
"""

with gr.Blocks(css="style.css") as demo:
    
    with gr.Column(elem_id="col-container"):
        
        gr.HTML(title)
        
        gpt_response = gr.Video(label="Talking Portrait response", elem_id="video_out")
             
    with gr.Column(elem_id="col-container-2"):
          
        record_input = gr.Audio(source="microphone",type="filepath", label="Audio input", show_label=True, elem_id="record_btn")
        whisper_tr = gr.Textbox(label="whisper english translation", elem_id="text_inp")

        send_btn = gr.Button("Send my request !")
    
        with gr.Group(elem_id="share-btn-container"):
                community_icon = gr.HTML(community_icon_html, visible=False)
                loading_icon = gr.HTML(loading_icon_html, visible=False)
                share_button = gr.Button("Share to community", elem_id="share-btn", visible=False)
    
    send_btn.click(infer, inputs=[record_input], outputs=[whisper_tr, gpt_response, share_button, community_icon, loading_icon])
    share_button.click(None, [], [], _js=share_js)

demo.queue(max_size=32, concurrency_count=20).launch(debug=True)