app.py

import gradio as gr
#import requests

from PIL import Image
import os
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):
 gpt_response = whisper_to_gpt(audio, "translate", fn_index=0)
 #print(gpt_response)
 audio_response = tts(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 portrait_link

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>
"""

css = '''
 #col-container, #col-container-2 {max-width: 510px; margin-left: auto; margin-right: auto;}
 a {text-decoration-line: underline; font-weight: 600;}
 div#record_btn > .mt-6 {
 margin-top: 0!important;
 }
 div#record_btn > .mt-6 button {
 width: 100%;
 height: 40px;
 }
 .footer {
 margin-bottom: 45px;
 margin-top: 10px;
 text-align: center;
 border-bottom: 1px solid #e5e5e5;
 }
 .footer>p {
 font-size: .8rem;
 display: inline-block;
 padding: 0 10px;
 transform: translateY(10px);
 background: white;
 }
 .dark .footer {
 border-color: #303030;
 }
 .dark .footer>p {
 background: #0b0f19;
 }
'''

with gr.Blocks(css=css) as demo:
 
 with gr.Column(elem_id="col-container"):
 
 gr.HTML(title)
 
 gpt_response = gr.Video(label="Talking Portrait response")
 
 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")

 send_btn = gr.Button("Send my request !")
 
 send_btn.click(infer, inputs=[record_input], outputs=[gpt_response])

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