app.py

import io, os, base64
from PIL import Image
import gradio as gr
import shortuuid
import numpy as np
from transformers import pipeline

asr = pipeline("automatic-speech-recognition")
latent = gr.Interface.load("spaces/multimodalart/latentdiffusion")


def text2image_latent(text, steps, width, height, images, diversity):
    print(text)
    results = latent(text, steps, width, height, images, diversity)
    image_paths = []
    for image in results[1]:
        image_str = image[0]
        image_str = image_str.replace("data:image/png;base64,","")
        decoded_bytes = base64.decodebytes(bytes(image_str, "utf-8"))
        img = Image.open(io.BytesIO(decoded_bytes))
        url = shortuuid.uuid()
        temp_dir = './tmp'
        if not os.path.exists(temp_dir):
            os.makedirs(temp_dir, exist_ok=True)
        image_path = f'{temp_dir}/{url}.png'
        img.save(f'{temp_dir}/{url}.png')
        image_paths.append(image_path)
    return(image_paths)


def speech_to_text(mic=None, file=None):
    if mic is not None:
        audio = mic
    elif file is not None:
        audio = file
    else:
        return "You must either provide a mic recording or a file"
    transcription = asr(audio)["text"]
    return transcription

    
with gr.Blocks() as demo: 
    gr.Markdown( """
    # 🎤 Sing or tell your story and let this Space ✨  visualize your story along 
    ## Inspired by this [tweet](https://twitter.com/karenxcheng/status/1516816114994454529?s=20&t=moq2vK5430JoerJXBTkIuA)
    ### Soon to be added:    
    - Near real time(streaming option)
    - Option playback of you audio relayed with video
     """)    
    with gr.Row():
        with gr.Column():
            audio_file =[
            gr.Audio(source="microphone", type="filepath", optional=True, label="Speak here..."),
            gr.Audio(source="upload", type="filepath", optional=True, label="Or if you want upload here...")]
            text = gr.Textbox(label="Text", placeholder="If you dont want to record or upload your voice you can input text here")
            with gr.Row():
                s2t = gr.Button("Speech to text go brrr")
        with gr.Column():
            steps = gr.inputs.Slider(label="Steps - more steps can increase quality but will take longer to     generate",default=1,maximum=50,minimum=1,step=1)
            width = gr.inputs.Slider(label="Width", default=256, step=32, maximum=256, minimum=32)
            height = gr.inputs.Slider(label="Height", default=256, step=32, maximum = 256, minimum=32)
            images = gr.inputs.Slider(label="Images - How many images you wish to generate", default=1, step=1, minimum=1, maximum=4)
            diversity = gr.inputs.Slider(label="Diversity scale - How different from one another you wish the images to be",default=15.0, minimum=1.0, maximum=15.0)
            gallery = gr.Gallery(label="Individual images")
            with gr.Row():
                get_image_latent = gr.Button("Generate Image go brr")         
        
    s2t.click(speech_to_text, inputs=audio_file, outputs=text)
    get_image_latent.click(text2image_latent, inputs=[text, steps, width, height, images, diversity], outputs=gallery)
      
demo.launch(enable_queue=True, debug=True)