File size: 2,233 Bytes
b5bfb01 fc53194 b5bfb01 fc53194 b5bfb01 fc53194 b5bfb01 |
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 os
import gradio as gr
import whisper
from gtts import gTTS
import io
import almlapi # Assuming AL/ML has a Python API library
# Set your AL/ML API key for authentication
os.environ["ALML_API_KEY"] = "701b35863e6d4a7b81bdcad2e6f3c880"
# Load the Whisper model for audio transcription
model = whisper.load_model("base")
# Function to process audio and interact with the AL/ML API
def process_audio(file_path):
try:
# Load and transcribe audio using Whisper
audio = whisper.load_audio(file_path)
result = model.transcribe(audio)
text = result["text"]
# Call OpenAI o1 model via AL/ML API for problem-solving
response = almlapi.call_o1(
api_key=os.environ.get("ALML_API_KEY"),
prompt=text,
model="o1" # Model name, adjust according to AL/ML documentation
)
# Extract the response message
response_message = response["generated_text"].strip()
# Convert response message to speech using gTTS
tts = gTTS(response_message)
response_audio_io = io.BytesIO()
tts.write_to_fp(response_audio_io) # Save the audio to BytesIO object
response_audio_io.seek(0)
# Save the audio file
with open("response.mp3", "wb") as audio_file:
audio_file.write(response_audio_io.getvalue())
# Return the response text and audio file path
return response_message, "response.mp3"
except Exception as e:
# Handle exceptions
return f"An error occurred: {e}", None
# Interface configurations (UI)
title = "Voice-to-Voice AI Chatbot with AL/ML API"
description = "Developed by [Adnan Tariq](https://www.linkedin.com/in/adnaantariq/) with ❤️"
article = "### Instructions\n1. Upload an audio file.\n2. Wait for the transcription.\n3. Listen to the chatbot's response."
# Gradio interface setup
iface = gr.Interface(
fn=process_audio,
inputs=gr.Audio(type="filepath"), # Upload audio via file path
outputs=[gr.Textbox(label="Response Text"), gr.Audio(label="Response Audio")],
live=True,
title=title,
description=description,
theme="dark",
article=article
)
# Launch the Gradio app
iface.launch()
|