Update app.py

app.py

@@ -3,28 +3,28 @@ from transformers import pipeline
 
 # function part
 # img2text
-def img2text(url):
-    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
+# def img2text(url):
+#     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):
-    text_to_story = pipeline("text-generation", model="isarth/distill_gpt2_story_generator",
-                            max_length=300,  # 增加最大生成长度
-                            min_length=100,  # 设置最小生成长度
-                            do_sample=True,  # 启用随机采样
-                            temperature=0.9,  # 控制随机性（0-1，越大越随机）
-                            top_k=50,  # 限制候选词数量
-                            top_p=0.95,  # 核采样参数
-                            repetition_penalty=1.2)
-    # story_text = ""   # to be completed
-    story_text = text_to_story(text)[0]["generated_text"]
-    return story_text
+# # text2story
+# def text2story(text):
+#     text_to_story = pipeline("text-generation", model="isarth/distill_gpt2_story_generator",
+#                             max_length=300,  # 增加最大生成长度
+#                             min_length=100,  # 设置最小生成长度
+#                             do_sample=True,  # 启用随机采样
+#                             temperature=0.9,  # 控制随机性（0-1，越大越随机）
+#                             top_k=50,  # 限制候选词数量
+#                             top_p=0.95,  # 核采样参数
+#                             repetition_penalty=1.2)
+#     # story_text = ""   # to be completed
+#     story_text = text_to_story(text)[0]["generated_text"]
+#     return story_text
 
 # text2audio
 def text2audio(story_text):
-    text_to_audio = pipeline("text-to-audio", model="facebook/musicgen-medium")
+    text_to_audio = ("text-to-speech", model="facebook/mms-tts-vie")
     audio_data = text_to_audio(story_text)
     return audio_data
     
@@ -43,22 +43,23 @@ if uploaded_file is not None:
 
     #Stage 1: Image to Text
     st.text('Processing img2text...')
-    scenario = img2text(uploaded_file.name)
-    st.write(scenario)
+    # scenario = img2text(uploaded_file.name)
+    # st.write(scenario)
 
     #Stage 2: Text to Story
     st.text('Generating a story...')
-    story = text2story(scenario)
+    # story = text2story(scenario)
+    story = "children playing in the park illustration and making fun of other characters. The film opens with a character named 'Sooja (Saabani) visiting his friend, Maithana Sarajan's house when they are not able to make any money from their loan by calling him at some stage which would be very difficult for them due lack on time-consuming situations; she had been told that it was going too far as if he could just get away once more without paying off the interest charges; her brother is worried about how much trouble Srinivasa will have between himself while studying abroad where one day someone gets hurt, then starts screaming "Hoopaam", so I'll take care." In another scene A girl tries desperately unsuccessfully vainly trying repeatedly all sorts tricks but no matter what happens before being caught alive or left behind, she becomes quite hysterical after seeing this movie again: It begins out loud singing like a child who wants nothing less than an old man dancing to dance outside a bank window, every little thing happening throughout its course takes place within seconds… At first sight however - although there isn't anything like that real action scenes such people often seem oblivious – it really shows how hard things can go through during childhood times! All these events show up together well into adulthood. This movie ends back towards home life at various stages showing great improvement including improvement along both lines until we know why many youngsters need those moments instead!! That said...the story concludes with...another series"
     st.write(story)
 
-    # #Stage 3: Story to Audio data
-    # st.text('Generating audio data...')
-    # audio_data =text2audio(story)
+    #Stage 3: Story to Audio data
+    st.text('Generating audio data...')
+    audio_data =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'])
+    # 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")