Spaces:
Runtime error
Runtime error
tomxxie
commited on
Commit
·
dd9bb13
1
Parent(s):
d9ae2cb
新增链接
Browse files- app.py +58 -26
- app_old.py +58 -26
app.py
CHANGED
|
@@ -158,11 +158,35 @@ def download_audio(input_wav_path):
|
|
| 158 |
else:
|
| 159 |
return None
|
| 160 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 161 |
# 创建 Gradio 界面
|
| 162 |
-
with gr.Blocks() as demo:
|
| 163 |
# 添加标题
|
| 164 |
gr.Markdown(
|
| 165 |
-
|
| 166 |
<div style="display: flex; align-items: center; justify-content: center; text-align: center;">
|
| 167 |
<h1 style="font-family: 'Arial', sans-serif; color: #014377; font-size: 32px; margin-bottom: 0; display: inline-block; vertical-align: middle;">
|
| 168 |
OSUM Speech Understanding Model Test
|
|
@@ -175,17 +199,17 @@ with gr.Blocks() as demo:
|
|
| 175 |
with gr.Row():
|
| 176 |
with gr.Column(scale=1):
|
| 177 |
audio_input = gr.Audio(label="Record", type="filepath")
|
| 178 |
-
with gr.Column(scale=1, min_width=300):
|
| 179 |
output_text = gr.Textbox(label="Output", lines=8, placeholder="The generated result will be displayed here...", interactive=False)
|
| 180 |
|
| 181 |
# 添加任务选择和自定义输入框
|
| 182 |
with gr.Row():
|
| 183 |
task_dropdown = gr.Dropdown(
|
| 184 |
label="Task",
|
| 185 |
-
choices=list(TASK_PROMPT_MAPPING.keys()) + ["Custom Task Prompt"],
|
| 186 |
value="ASR (Automatic Speech Recognition)"
|
| 187 |
)
|
| 188 |
-
custom_prompt_input = gr.Textbox(label="Custom Task Prompt", placeholder="Please enter a custom task prompt...", visible=False)
|
| 189 |
|
| 190 |
# 添加按钮(下载按钮在左边,开始处理按钮在右边)
|
| 191 |
with gr.Row():
|
|
@@ -205,20 +229,24 @@ with gr.Blocks() as demo:
|
|
| 205 |
save_button = gr.Button("Submit Feedback", variant="secondary")
|
| 206 |
|
| 207 |
# 添加底部内容
|
| 208 |
-
|
| 209 |
-
|
| 210 |
-
|
| 211 |
-
|
| 212 |
-
|
| 213 |
-
|
| 214 |
-
|
| 215 |
-
|
| 216 |
-
|
| 217 |
-
|
| 218 |
-
|
| 219 |
-
|
| 220 |
-
|
| 221 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 222 |
|
| 223 |
# 绑定事件
|
| 224 |
def show_confirmation(output_res, input_wav_path, input_prompt):
|
|
@@ -229,12 +257,16 @@ with gr.Blocks() as demo:
|
|
| 229 |
return gr.update(visible=False)
|
| 230 |
|
| 231 |
def handle_submit(input_wav_path, task_choice, custom_prompt):
|
| 232 |
-
|
| 233 |
-
|
| 234 |
-
|
| 235 |
-
|
| 236 |
-
|
| 237 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 238 |
|
| 239 |
# 当任务选择框的值发生变化时,更新自定义输入框的可见性
|
| 240 |
task_dropdown.change(
|
|
@@ -256,7 +288,7 @@ with gr.Blocks() as demo:
|
|
| 256 |
download_button.click(
|
| 257 |
fn=download_audio,
|
| 258 |
inputs=[audio_input],
|
| 259 |
-
outputs=[download_button]
|
| 260 |
)
|
| 261 |
|
| 262 |
save_button.click(
|
|
|
|
| 158 |
else:
|
| 159 |
return None
|
| 160 |
|
| 161 |
+
# 自定义 CSS 样式
|
| 162 |
+
CSS = """
|
| 163 |
+
.custom-footer {
|
| 164 |
+
position: fixed;
|
| 165 |
+
bottom: 20px; /* 距离页面底部的距离 */
|
| 166 |
+
left: 50%;
|
| 167 |
+
transform: translateX(-50%);
|
| 168 |
+
display: flex;
|
| 169 |
+
align-items: center;
|
| 170 |
+
justify-content: center;
|
| 171 |
+
gap: 20px;
|
| 172 |
+
text-align: center;
|
| 173 |
+
font-weight: bold;
|
| 174 |
+
padding-bottom: 20px; /* 在底部添加额外的间距 */
|
| 175 |
+
}
|
| 176 |
+
.custom-footer p {
|
| 177 |
+
margin: 0;
|
| 178 |
+
}
|
| 179 |
+
.custom-footer img {
|
| 180 |
+
height: 80px;
|
| 181 |
+
width: auto;
|
| 182 |
+
}
|
| 183 |
+
"""
|
| 184 |
+
|
| 185 |
# 创建 Gradio 界面
|
| 186 |
+
with gr.Blocks(css=CSS) as demo:
|
| 187 |
# 添加标题
|
| 188 |
gr.Markdown(
|
| 189 |
+
"""
|
| 190 |
<div style="display: flex; align-items: center; justify-content: center; text-align: center;">
|
| 191 |
<h1 style="font-family: 'Arial', sans-serif; color: #014377; font-size: 32px; margin-bottom: 0; display: inline-block; vertical-align: middle;">
|
| 192 |
OSUM Speech Understanding Model Test
|
|
|
|
| 199 |
with gr.Row():
|
| 200 |
with gr.Column(scale=1):
|
| 201 |
audio_input = gr.Audio(label="Record", type="filepath")
|
| 202 |
+
with gr.Column(scale=1, min_width=300):
|
| 203 |
output_text = gr.Textbox(label="Output", lines=8, placeholder="The generated result will be displayed here...", interactive=False)
|
| 204 |
|
| 205 |
# 添加任务选择和自定义输入框
|
| 206 |
with gr.Row():
|
| 207 |
task_dropdown = gr.Dropdown(
|
| 208 |
label="Task",
|
| 209 |
+
choices=list(TASK_PROMPT_MAPPING.keys()) + ["Custom Task Prompt"],
|
| 210 |
value="ASR (Automatic Speech Recognition)"
|
| 211 |
)
|
| 212 |
+
custom_prompt_input = gr.Textbox(label="Custom Task Prompt", placeholder="Please enter a custom task prompt...", visible=False)
|
| 213 |
|
| 214 |
# 添加按钮(下载按钮在左边,开始处理按钮在右边)
|
| 215 |
with gr.Row():
|
|
|
|
| 229 |
save_button = gr.Button("Submit Feedback", variant="secondary")
|
| 230 |
|
| 231 |
# 添加底部内容
|
| 232 |
+
gr.HTML(
|
| 233 |
+
f"""
|
| 234 |
+
<div class="custom-footer">
|
| 235 |
+
<div>
|
| 236 |
+
<p>
|
| 237 |
+
<a href="http://www.nwpu-aslp.org/" target="_blank">Audio, Speech and Language Processing Group (ASLP@NPU)</a>,
|
| 238 |
+
</p>
|
| 239 |
+
<p>
|
| 240 |
+
Northwestern Polytechnical University
|
| 241 |
+
</p>
|
| 242 |
+
<p>
|
| 243 |
+
<a href="https://github.com/ASLP-lab/OSUM" target="_blank">GitHub</a>
|
| 244 |
+
</p>
|
| 245 |
+
</div>
|
| 246 |
+
<img src="data:image/png;base64,{encoded_string}" alt="OSUM Logo">
|
| 247 |
+
</div>
|
| 248 |
+
"""
|
| 249 |
+
)
|
| 250 |
|
| 251 |
# 绑定事件
|
| 252 |
def show_confirmation(output_res, input_wav_path, input_prompt):
|
|
|
|
| 257 |
return gr.update(visible=False)
|
| 258 |
|
| 259 |
def handle_submit(input_wav_path, task_choice, custom_prompt):
|
| 260 |
+
try:
|
| 261 |
+
if task_choice == "Custom Task Prompt":
|
| 262 |
+
input_prompt = custom_prompt
|
| 263 |
+
else:
|
| 264 |
+
input_prompt = TASK_PROMPT_MAPPING.get(task_choice, "未知任务类型")
|
| 265 |
+
output_res = do_decode(input_wav_path, input_prompt)
|
| 266 |
+
return output_res
|
| 267 |
+
except Exception as e:
|
| 268 |
+
print(f"Error in handle_submit: {e}")
|
| 269 |
+
return "Error occurred. Please check the input."
|
| 270 |
|
| 271 |
# 当任务选择框的值发生变化时,更新自定义输入框的可见性
|
| 272 |
task_dropdown.change(
|
|
|
|
| 288 |
download_button.click(
|
| 289 |
fn=download_audio,
|
| 290 |
inputs=[audio_input],
|
| 291 |
+
outputs=[download_button]
|
| 292 |
)
|
| 293 |
|
| 294 |
save_button.click(
|
app_old.py
CHANGED
|
@@ -158,11 +158,35 @@ def download_audio(input_wav_path):
|
|
| 158 |
else:
|
| 159 |
return None
|
| 160 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 161 |
# 创建 Gradio 界面
|
| 162 |
-
with gr.Blocks() as demo:
|
| 163 |
# 添加标题
|
| 164 |
gr.Markdown(
|
| 165 |
-
|
| 166 |
<div style="display: flex; align-items: center; justify-content: center; text-align: center;">
|
| 167 |
<h1 style="font-family: 'Arial', sans-serif; color: #014377; font-size: 32px; margin-bottom: 0; display: inline-block; vertical-align: middle;">
|
| 168 |
OSUM Speech Understanding Model Test
|
|
@@ -175,17 +199,17 @@ with gr.Blocks() as demo:
|
|
| 175 |
with gr.Row():
|
| 176 |
with gr.Column(scale=1):
|
| 177 |
audio_input = gr.Audio(label="Record", type="filepath")
|
| 178 |
-
with gr.Column(scale=1, min_width=300):
|
| 179 |
output_text = gr.Textbox(label="Output", lines=8, placeholder="The generated result will be displayed here...", interactive=False)
|
| 180 |
|
| 181 |
# 添加任务选择和自定义输入框
|
| 182 |
with gr.Row():
|
| 183 |
task_dropdown = gr.Dropdown(
|
| 184 |
label="Task",
|
| 185 |
-
choices=list(TASK_PROMPT_MAPPING.keys()) + ["Custom Task Prompt"],
|
| 186 |
value="ASR (Automatic Speech Recognition)"
|
| 187 |
)
|
| 188 |
-
custom_prompt_input = gr.Textbox(label="Custom Task Prompt", placeholder="Please enter a custom task prompt...", visible=False)
|
| 189 |
|
| 190 |
# 添加按钮(下载按钮在左边,开始处理按钮在右边)
|
| 191 |
with gr.Row():
|
|
@@ -205,20 +229,24 @@ with gr.Blocks() as demo:
|
|
| 205 |
save_button = gr.Button("Submit Feedback", variant="secondary")
|
| 206 |
|
| 207 |
# 添加底部内容
|
| 208 |
-
|
| 209 |
-
|
| 210 |
-
|
| 211 |
-
|
| 212 |
-
|
| 213 |
-
|
| 214 |
-
|
| 215 |
-
|
| 216 |
-
|
| 217 |
-
|
| 218 |
-
|
| 219 |
-
|
| 220 |
-
|
| 221 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 222 |
|
| 223 |
# 绑定事件
|
| 224 |
def show_confirmation(output_res, input_wav_path, input_prompt):
|
|
@@ -229,12 +257,16 @@ with gr.Blocks() as demo:
|
|
| 229 |
return gr.update(visible=False)
|
| 230 |
|
| 231 |
def handle_submit(input_wav_path, task_choice, custom_prompt):
|
| 232 |
-
|
| 233 |
-
|
| 234 |
-
|
| 235 |
-
|
| 236 |
-
|
| 237 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
| 238 |
|
| 239 |
# 当任务选择框的值发生变化时,更新自定义输入框的可见性
|
| 240 |
task_dropdown.change(
|
|
@@ -256,7 +288,7 @@ with gr.Blocks() as demo:
|
|
| 256 |
download_button.click(
|
| 257 |
fn=download_audio,
|
| 258 |
inputs=[audio_input],
|
| 259 |
-
outputs=[download_button]
|
| 260 |
)
|
| 261 |
|
| 262 |
save_button.click(
|
|
|
|
| 158 |
else:
|
| 159 |
return None
|
| 160 |
|
| 161 |
+
# 自定义 CSS 样式
|
| 162 |
+
CSS = """
|
| 163 |
+
.custom-footer {
|
| 164 |
+
position: fixed;
|
| 165 |
+
bottom: 20px; /* 距离页面底部的距离 */
|
| 166 |
+
left: 50%;
|
| 167 |
+
transform: translateX(-50%);
|
| 168 |
+
display: flex;
|
| 169 |
+
align-items: center;
|
| 170 |
+
justify-content: center;
|
| 171 |
+
gap: 20px;
|
| 172 |
+
text-align: center;
|
| 173 |
+
font-weight: bold;
|
| 174 |
+
padding-bottom: 20px; /* 在底部添加额外的间距 */
|
| 175 |
+
}
|
| 176 |
+
.custom-footer p {
|
| 177 |
+
margin: 0;
|
| 178 |
+
}
|
| 179 |
+
.custom-footer img {
|
| 180 |
+
height: 80px;
|
| 181 |
+
width: auto;
|
| 182 |
+
}
|
| 183 |
+
"""
|
| 184 |
+
|
| 185 |
# 创建 Gradio 界面
|
| 186 |
+
with gr.Blocks(css=CSS) as demo:
|
| 187 |
# 添加标题
|
| 188 |
gr.Markdown(
|
| 189 |
+
"""
|
| 190 |
<div style="display: flex; align-items: center; justify-content: center; text-align: center;">
|
| 191 |
<h1 style="font-family: 'Arial', sans-serif; color: #014377; font-size: 32px; margin-bottom: 0; display: inline-block; vertical-align: middle;">
|
| 192 |
OSUM Speech Understanding Model Test
|
|
|
|
| 199 |
with gr.Row():
|
| 200 |
with gr.Column(scale=1):
|
| 201 |
audio_input = gr.Audio(label="Record", type="filepath")
|
| 202 |
+
with gr.Column(scale=1, min_width=300):
|
| 203 |
output_text = gr.Textbox(label="Output", lines=8, placeholder="The generated result will be displayed here...", interactive=False)
|
| 204 |
|
| 205 |
# 添加任务选择和自定义输入框
|
| 206 |
with gr.Row():
|
| 207 |
task_dropdown = gr.Dropdown(
|
| 208 |
label="Task",
|
| 209 |
+
choices=list(TASK_PROMPT_MAPPING.keys()) + ["Custom Task Prompt"],
|
| 210 |
value="ASR (Automatic Speech Recognition)"
|
| 211 |
)
|
| 212 |
+
custom_prompt_input = gr.Textbox(label="Custom Task Prompt", placeholder="Please enter a custom task prompt...", visible=False)
|
| 213 |
|
| 214 |
# 添加按钮(下载按钮在左边,开始处理按钮在右边)
|
| 215 |
with gr.Row():
|
|
|
|
| 229 |
save_button = gr.Button("Submit Feedback", variant="secondary")
|
| 230 |
|
| 231 |
# 添加底部内容
|
| 232 |
+
gr.HTML(
|
| 233 |
+
f"""
|
| 234 |
+
<div class="custom-footer">
|
| 235 |
+
<div>
|
| 236 |
+
<p>
|
| 237 |
+
<a href="http://www.nwpu-aslp.org/" target="_blank">Audio, Speech and Language Processing Group (ASLP@NPU)</a>,
|
| 238 |
+
</p>
|
| 239 |
+
<p>
|
| 240 |
+
Northwestern Polytechnical University
|
| 241 |
+
</p>
|
| 242 |
+
<p>
|
| 243 |
+
<a href="https://github.com/ASLP-lab/OSUM" target="_blank">GitHub</a>
|
| 244 |
+
</p>
|
| 245 |
+
</div>
|
| 246 |
+
<img src="data:image/png;base64,{encoded_string}" alt="OSUM Logo">
|
| 247 |
+
</div>
|
| 248 |
+
"""
|
| 249 |
+
)
|
| 250 |
|
| 251 |
# 绑定事件
|
| 252 |
def show_confirmation(output_res, input_wav_path, input_prompt):
|
|
|
|
| 257 |
return gr.update(visible=False)
|
| 258 |
|
| 259 |
def handle_submit(input_wav_path, task_choice, custom_prompt):
|
| 260 |
+
try:
|
| 261 |
+
if task_choice == "Custom Task Prompt":
|
| 262 |
+
input_prompt = custom_prompt
|
| 263 |
+
else:
|
| 264 |
+
input_prompt = TASK_PROMPT_MAPPING.get(task_choice, "未知任务类型")
|
| 265 |
+
output_res = do_decode(input_wav_path, input_prompt)
|
| 266 |
+
return output_res
|
| 267 |
+
except Exception as e:
|
| 268 |
+
print(f"Error in handle_submit: {e}")
|
| 269 |
+
return "Error occurred. Please check the input."
|
| 270 |
|
| 271 |
# 当任务选择框的值发生变化时,更新自定义输入框的可见性
|
| 272 |
task_dropdown.change(
|
|
|
|
| 288 |
download_button.click(
|
| 289 |
fn=download_audio,
|
| 290 |
inputs=[audio_input],
|
| 291 |
+
outputs=[download_button]
|
| 292 |
)
|
| 293 |
|
| 294 |
save_button.click(
|