Update app.py

app.py

@@ -1,66 +1,61 @@
 # import part
 import streamlit as st
 from transformers import pipeline
-from gtts import gTTS
-import os
+import torch
+import soundfile as sf
+import numpy as np
 
 # function part
 # img2text
 def img2text(url):
-    image_to_text_model = pipeline("image-to-text", 
-                                   model="Salesforce/blip-image-captioning-base")
+    image_to_text_model = pipeline("image-to-text", model="Salesforce/blip-image-captioning-base")
     text = image_to_text_model(url)[0]["generated_text"]
     return text
 
 # text2story
 def text2story(text):
-    story_text = pipeline("text-generation", model="perplexity-ai/r1-1776", trust_remote_code=True)  # to be completed
-    return story_text
+    story_text_model = pipeline("text-generation", model="Qwen/QwQ-32B")
+    story = story_text_model(text, max_length=150)[0]['generated_text']
+    return story
 
 # text2audio
 def text2audio(story_text):
-    # Convert text to audio using gTTS
-    tts = gTTS(story_text, lang="en")
-    audio_file = "story_audio.wav"
-    tts.save(audio_file)
-    return audio_file
+    # Here we will use a text-to-speech model from Hugging Face
+    tts_model = pipeline("text-to-speech", model="tts_models/en/ljspeech/tacotron2")
+    audio_data = tts_model(story_text)
+    
+    # Save audio to a file
+    audio_filename = "story_audio.wav"
+    sf.write(audio_filename, audio_data['audio'], audio_data['sampling_rate'])
+    
+    return audio_filename
 
 # main part
-st.set_page_config(page_title="Your Image to Audio Story",
-                   page_icon="🦜") # prepare configuration
+st.set_page_config(page_title="Your Image to Audio Story", page_icon="🦜")
 st.header("Turn Your Image to Audio Story")
-
-# Upload image
 uploaded_file = st.file_uploader("Select an Image...")
 
-# If it is none, skip all the following things
 if uploaded_file is not None:
-    print(uploaded_file)
     bytes_data = uploaded_file.getvalue()
     with open(uploaded_file.name, "wb") as file:
         file.write(bytes_data)
 
-    st.image(uploaded_file, caption="Uploaded Image",
-             use_column_width=True)
+    st.image(uploaded_file, caption="Uploaded Image", use_column_width=True)
 
-    #Stage 1: Image to Text
+    # Stage 1: Image to Text
     st.text('Processing img2text...')
     scenario = img2text(uploaded_file.name)
     st.write(scenario)
 
-    #Stage 2: Text to Story
+    # Stage 2: Text to Story
     st.text('Generating a story...')
     story = text2story(scenario)
     st.write(story)
 
-    #Stage 3: Story to Audio data
+    # Stage 3: Story to Audio data
     st.text('Generating audio data...')
-    audio_data =text2audio(story)
+    audio_filename = text2audio(story)
 
     # Play button
     if st.button("Play Audio"):
-        st.audio(audio_data['audio'],
-                    format="audio/wav",
-                    start_time=0,
-                    sample_rate = audio_data['sampling_rate'])
-        st.audio("kids_playing_audio.wav")
+        st.audio(audio_filename, format="audio/wav")