app.py

from transformers import pipeline
from huggingface_hub import login
from diffusers import StableDiffusionPipeline
import gradio as gr
import torch
import transformers
# Set Hugging Face token
hf_token = "your_huggingface_token_here"  # Replace this with your token
login(hf_token)

# Load Hugging Face models
speech_to_text = pipeline("automatic-speech-recognition", model="openai/whisper-base")

# Load Stable Diffusion model using diffusers
text_to_image = StableDiffusionPipeline.from_pretrained(
    "runwayml/stable-diffusion-v1-5", torch_dtype=torch.float16
).to("cuda" if torch.cuda.is_available() else "cpu")

# Speech-to-text function
def transcribe_audio(audio_file):
    try:
        result = speech_to_text(audio_file)
        transcription = result["text"]
        return transcription
    except Exception as e:
        return f"Error in transcription: {str(e)}"

# Text-to-image function
def generate_image_from_text(text):
    try:
        image = text_to_image(text).images[0]  # Generate one image
        return image
    except Exception as e:
        return f"Error in image generation: {str(e)}"

# Combined processing function
def process_audio_and_generate_image(audio_file):
    transcription = transcribe_audio(audio_file)
    if "Error" in transcription:
        return None, transcription

    image = generate_image_from_text(transcription)
    if isinstance(image, str) and "Error" in image:
        return None, image

    return image, transcription

# Gradio interface
iface = gr.Interface(
    fn=process_audio_and_generate_image,
    inputs=gr.Audio(type="filepath", label="Upload audio file (WAV/MP3)"),
    outputs=[
        gr.Image(label="Generated Image"),
        gr.Textbox(label="Transcription")
    ],
    title="Speech-to-Text and Image Generation",
    description="Upload an audio file to transcribe speech to text, and then generate an image based on the transcription.",
)

# Launch the interface
iface.launch(share=True)