Spaces:
Sleeping
Sleeping
File size: 2,790 Bytes
dbb5b1d 5c75cf0 91773b9 5c75cf0 91773b9 5c75cf0 91773b9 5c75cf0 91773b9 5c75cf0 91773b9 5c75cf0 91773b9 5c75cf0 91773b9 dbb5b1d 5c75cf0 91773b9 |
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 79 80 81 82 83 84 85 86 87 88 89 90 |
from smolagents import load_tool, CodeAgent, HfApiModel, DuckDuckGoSearchTool
#from dotenv import load_dotenv
from smolagents import CodeAgent, DuckDuckGoSearchTool, HfApiModel, ManagedAgent, VisitWebpageTool, tool
model = HfApiModel()
search_tool = DuckDuckGoSearchTool()
visit_webpage_tool = VisitWebpageTool()
agent = CodeAgent(
tools=[search_tool, visit_webpage_tool],
model=model,
additional_authorized_imports=['requests', 'bs4', 'pandas', 'gradio', 'concurrent.futures', 'csv', 'json']
)
"""Deploying AI Voice Chatbot Gradio App."""
import gradio as gr
from typing import Tuple
from utils import (
TextGenerationPipeline,
from_en_translation,
html_audio_autoplay,
stt,
to_en_translation,
tts,
tts_to_bytesio,
)
max_answer_length = 100
desired_language = "de"
response_generator_pipe = TextGenerationPipeline(max_length=max_answer_length)
def main(audio: object) -> Tuple[str, str, str, object]:
"""Calls functions for deploying Gradio app.
It responds both verbally and in text
by taking voice input from the user.
Args:
audio (object): Recorded speech of the user.
Returns:
tuple containing:
- user_speech_text (str): Recognized speech.
- bot_response_de (str): Translated answer of the bot.
- bot_response_en (str): Bot's original answer.
- html (object): Autoplayer for bot's speech.
"""
user_speech_text = stt(audio, desired_language)
translated_text = to_en_translation(user_speech_text, desired_language)
#TODO call the agent
# bot_response_en = response_generator_pipe(translated_text)
bot_response_en = agent.run(translated_text)
###
bot_response_de = from_en_translation(bot_response_en, desired_language)
bot_voice = tts(bot_response_de, desired_language)
bot_voice_bytes = tts_to_bytesio(bot_voice)
html = html_audio_autoplay(bot_voice_bytes)
return user_speech_text, bot_response_de, bot_response_en, html
# Define the Gradio interface
with gr.Blocks() as demo:
gr.Markdown("## AI Voice Chatbot")
with gr.Row():
audio_input = gr.Audio(type="filepath", label="Speak or Upload Audio")
submit_btn = gr.Button("Submit")
with gr.Row():
user_speech_text = gr.Textbox(label="You said:", interactive=False)
bot_response_de = gr.Textbox(label="AI said (in German):", interactive=False)
bot_response_en = gr.Textbox(label="AI said (in English):", interactive=False)
html_output = gr.HTML()
# Connect the function to the components
submit_btn.click(
fn=main,
inputs=[audio_input],
outputs=[user_speech_text, bot_response_de, bot_response_en, html_output],
)
# Launch the Gradio app
demo.launch(debug=True)
|