Spaces:
Build error
Build error
File size: 1,715 Bytes
fcfc5d9 694ecc6 fcfc5d9 694ecc6 fcfc5d9 694ecc6 |
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 69 70 71 72 73 74 75 76 77 78 |
"""PUBLIC
Simple demo app.
Copyright PolyAI Limited.
"""
import time
from pathlib import Path
import gradio as gr
from transformer_infer import PhemeClient, parse_arguments
# TODO
VOICE_OPTIONS = [
"male_voice",
"POD1000000004_S0000246",
"POD1000000018_S0000253",
"POD1000000048_S0000035",
"YOU1000000006_S0000051",
"YOU1000000044_S0000798",
"empress",
]
args = parse_arguments()
model = PhemeClient(args)
def inference(
text,
voice,
top_k,
temperature
):
with open("PhemeVoice.log", "a") as f:
f.write(f"{voice}: {text} \n")
start_time = time.time()
data = model.infer(
text, voice, top_k=top_k, temperature=temperature)
samplerate = 16_000
print("Time taken: ", time.time() - start_time)
yield (samplerate, data)
def main():
title = "Pheme"
description = """Model can generate a variety of conversational voices."""
text = gr.Textbox(
lines=3,
value="Our property has several wedding venues.",
label="Text",
)
voice = gr.Dropdown(
VOICE_OPTIONS, value="empress", label="Select voice:", type="value"
)
temperature = gr.Slider(minimum=.3, maximum=1.5, value=0.7, step=0.05)
top_k = gr.Slider(minimum=10, maximum=250, value=210)
output_audio = gr.Audio(label="audio:", autoplay=True)
interface = gr.Interface(
fn=inference,
inputs=[
text,
voice,
top_k,
temperature,
],
title=title,
description=description,
outputs=[output_audio],
)
interface.queue().launch(share=True)
if __name__ == "__main__":
main()
|