File size: 1,919 Bytes
7572555 340d310 7572555 340d310 7572555 340d310 7572555 340d310 7572555 340d310 7572555 340d310 7572555 1242299 6c305c0 7572555 60f71bf 7572555 7d1064d 1e7779c 84ddd8e 1e7779c 7572555 6b8c3fc 7572555 2c65526 7572555 2c65526 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 |
import gradio as gr
from gradio_client import Client, handle_file
def get_speech(text, voice):
client = Client("collabora/WhisperSpeech")
result = client.predict(
multilingual_text=text,
speaker_audio=voice,
speaker_url="",
cps=14,
api_name="/whisper_speech_demo"
)
print(result)
return result
def get_dreamtalk(image_in, speech):
client = Client("fffiloni/dreamtalk")
result = client.predict(
audio_input=handle_file(speech),
image_path=handle_file(image_in),
emotional_style="M030_front_neutral_level1_001.mat",
api_name="/infer"
)
print(result)
return result['video']
def pipe (text, voice, image_in):
speech = get_speech(text, voice)
try:
video = get_dreamtalk(image_in, speech)
except:
raise gr.Error('An error occurred while loading DreamTalk: Image may not contain any face')
return video
with gr.Blocks() as demo:
with gr.Column():
gr.HTML("""
<h2 style="text-align: center;">
Whisper Speech X Dreamtalk
</h2>
<p style="text-align: center;"></p>
""")
with gr.Row():
with gr.Column():
image_in = gr.Image(label="Portrait IN", type="filepath", value="./einstein.jpg")
with gr.Column():
voice = gr.Audio(type="filepath", label="Upload or Record Speaker audio (Optional voice cloning)")
text = gr.Textbox(label="text")
submit_btn = gr.Button('Submit')
with gr.Column():
video_o = gr.Video(label="Video result")
submit_btn.click(
fn = pipe,
inputs = [
text, voice, image_in
],
outputs = [
video_o
],
concurrency_limit = 3
)
demo.queue(max_size=10).launch(show_error=True, show_api=False) |