Update app.py

app.py

@@ -4,11 +4,11 @@ from transformers import pipeline
 from gtts import gTTS
 import os
 
-# 加载 Hugging Face 模型
+# 加載 Hugging Face 模型
 image_to_text_model = pipeline("image-to-text", model="Salesforce/blip-image-captioning-base")
 story_generator = pipeline("text-generation", model="facebook/opt-1.3b")
 
-# 图片 → 文字（生成描述）
+# 圖片 → 文字（生成描述）
 def img2text(image_path):
     text = image_to_text_model(image_path)[0]["generated_text"]
     return text
@@ -19,14 +19,20 @@ def text2story(text):
     story = story_generator(prompt, max_length=250, do_sample=True, temperature=0.8, top_p=0.9, repetition_penalty=1.2, truncation=True)[0]['generated_text']
     return story
 
-# 故事 → 语音（TTS）
+# 故事 → 語音（TTS）
 def text2audio_gtts(story_text, filename="story.mp3"):
+    # 避免文件冲突
     if os.path.exists(filename):
         os.remove(filename)
 
-    story_text = story_text[:500]  # 限制 TTS 文本长度
+    # 限制 TTS 文本長度
+    max_chars = 500  # gTTS 可能不支持過長文本
+    story_text = story_text[:max_chars]
+
+    # 生成语音
     tts = gTTS(text=story_text, lang="en")
     tts.save(filename)
+
     return filename
 
 # Streamlit Web UI
@@ -36,26 +42,33 @@ st.header("📖 AI Storyteller: Turn Your Image into a Story with Audio")
 uploaded_file = st.file_uploader("Upload an Image...", type=["jpg", "png"])
 
 if uploaded_file:
+    # 保存圖片到本地
     image_path = "uploaded_image.jpg"
     with open(image_path, "wb") as f:
         f.write(uploaded_file.getbuffer())
 
+    # 讀取並顯示圖片
     image = Image.open(image_path)
     st.image(image, caption="Uploaded Image", use_column_width=True)
 
+    # 生成圖片描述
     st.text("🔍 Generating image caption...")
-    caption = img2text(image_path)
+    caption = img2text(image_path)  # 這裏改成文件路徑
     st.write("**Image Description:**", caption)
 
+    # 生成故事
     st.text("📝 Generating story...")
     story = text2story(caption)
     st.write("**Generated Story:**")
     st.write(story)
 
+    # 生成語音
     st.text("🔊 Generating audio...")
     audio_file = text2audio_gtts(story)
 
+    # 播放音頻
     st.audio(audio_file, format="audio/mp3")
 
+    # 下載按钮
     with open(audio_file, "rb") as file:
         st.download_button("📥 Download Audio", file, file_name="story.mp3")