Spaces:
Runtime error
Runtime error
File size: 1,672 Bytes
bd3e2e4 3b38860 bd3e2e4 3b38860 bd3e2e4 82f0eab bd3e2e4 82f0eab bd3e2e4 cc06fc8 3b38860 bd3e2e4 3b38860 bd3e2e4 3b38860 bd3e2e4 |
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 |
import gradio as gr
from faster_whisper import WhisperModel
model = WhisperModel("tiny")
def generate_response(correction_intensity,
language_level,
buddy_personality,
language_choice,
user_query_audio
):
# Convert input audio to text
language_codes = {'English':'en',
'Urdu':'ur',
'Japanese':'ja'}
user_query_transcribed_segments, info = model.transcribe(
audio=user_query_audio,
language=language_codes[language_choice]
)
user_query_transcribed = list(user_query_transcribed_segments)[0].text.strip()
# Ask llm for response to text
# Convert llm response to audio
# Return converted llm response
return user_query_transcribed
demo = gr.Interface(
fn=generate_response,
inputs=[
gr.Slider(
minimum=1,
maximum=5,
step=1,
label='Grammar Correction Intensity'
),
gr.Dropdown(
choices=['Beginner', 'Intermediate', 'Advanced'],
label='Language Level'),
gr.Dropdown(
choices=['Formal Teacher', 'Flirty Friend', 'Sarcastic Bro'],
label='Language Buddy Personality'),
gr.Dropdown(
choices=['English', 'Urdu', 'Japanese'],
label='Language Choice'),
gr.Audio(
sources='microphone',
show_download_button=True,
type='filepath'
)],
outputs=[
gr.Textbox(label='AI Buddy Response')
],
title="AI Language Buddy"
)
demo.launch() |