Update app.py
Browse files
app.py
CHANGED
|
@@ -7,9 +7,6 @@ import torch
|
|
| 7 |
st.set_page_config(page_title="Image to Audio Story", page_icon="🦜")
|
| 8 |
|
| 9 |
def extract_image_caption(image_data):
|
| 10 |
-
"""
|
| 11 |
-
利用预训练模型从图像中提取描述性文字。
|
| 12 |
-
"""
|
| 13 |
img_obj = Image.open(image_data)
|
| 14 |
caption_pipeline = pipeline(
|
| 15 |
"image-to-text",
|
|
@@ -20,9 +17,6 @@ caption_text = caption_results[0]['generated_text']
|
|
| 20 |
return caption_text
|
| 21 |
|
| 22 |
def compose_story_from_caption(caption_detail):
|
| 23 |
-
"""
|
| 24 |
-
根据图像描述创作一篇充满创意的儿童故事。
|
| 25 |
-
"""
|
| 26 |
story_pipeline = pipeline(
|
| 27 |
"text-generation",
|
| 28 |
model="Qwen/Qwen2-1.5B",
|
|
@@ -43,9 +37,6 @@ story = story_text.strip()
|
|
| 43 |
return story
|
| 44 |
|
| 45 |
def convert_text_to_audio(text_content, audio_path="output.mp3"):
|
| 46 |
-
"""
|
| 47 |
-
将文本转换为音频文件。
|
| 48 |
-
"""
|
| 49 |
tts_engine = gTTS(text=text_content, lang="en")
|
| 50 |
tts_engine.save(audio_path)
|
| 51 |
return audio_path
|
|
|
|
| 7 |
st.set_page_config(page_title="Image to Audio Story", page_icon="🦜")
|
| 8 |
|
| 9 |
def extract_image_caption(image_data):
|
|
|
|
|
|
|
|
|
|
| 10 |
img_obj = Image.open(image_data)
|
| 11 |
caption_pipeline = pipeline(
|
| 12 |
"image-to-text",
|
|
|
|
| 17 |
return caption_text
|
| 18 |
|
| 19 |
def compose_story_from_caption(caption_detail):
|
|
|
|
|
|
|
|
|
|
| 20 |
story_pipeline = pipeline(
|
| 21 |
"text-generation",
|
| 22 |
model="Qwen/Qwen2-1.5B",
|
|
|
|
| 37 |
return story
|
| 38 |
|
| 39 |
def convert_text_to_audio(text_content, audio_path="output.mp3"):
|
|
|
|
|
|
|
|
|
|
| 40 |
tts_engine = gTTS(text=text_content, lang="en")
|
| 41 |
tts_engine.save(audio_path)
|
| 42 |
return audio_path
|