update

README.md

@@ -16,6 +16,7 @@ Check out the configuration reference at https://huggingface.co/docs/hub/spaces-
 docker build -t asr:v20240510_1734 .
 
 docker run -itd --name ASR \
+--network host \
 -v /data/tianxing/PycharmProjects/asr/data/:/home/user/app/data/ \
 -v /data/tianxing/PycharmProjects/asr/pretrained_models/:/home/user/app/pretrained_models/ \
 asr:v20240510_1734 /bin/bash

main.py

@@ -59,6 +59,16 @@ def process_uploaded_file(language: str,
     return "Dummy", build_html_output("Dummy")
 
 
+# css style is copied from
+# https://huggingface.co/spaces/alphacep/asr/blob/main/app.py#L113
+css = """
+.result {display:flex;flex-direction:column}
+.result_item {padding:15px;margin-bottom:8px;border-radius:15px;width:100%}
+.result_item_success {background-color:mediumaquamarine;color:white;align-self:start}
+.result_item_error {background-color:#ff7070;color:white;align-self:start}
+"""
+
+
 def main():
     title = "# Automatic Speech Recognition with Next-gen Kaldi"
 
@@ -68,40 +78,39 @@ def main():
         "Chinese": ["None"]
     }
 
-    # components
-    language_radio = gr.Radio(
-        label="Language",
-        choices=language_choices,
-        value=language_choices[0],
-    )
-    model_dropdown = gr.Dropdown(
-        choices=language_to_models[language_choices[0]],
-        label="Select a model",
-        value=language_to_models[language_choices[0]][0],
-    )
-    decoding_method_radio = gr.Radio(
-        label="Decoding method",
-        choices=["greedy_search", "modified_beam_search"],
-        value="greedy_search",
-    )
-    num_active_paths_slider = gr.Slider(
-        minimum=1,
-        value=4,
-        step=1,
-        label="Number of active paths for modified_beam_search",
-    )
-    punct_radio = gr.Radio(
-        label="Whether to add punctuation (Only for Chinese and English)",
-        choices=["Yes", "No"],
-        value="Yes",
-    )
-
     # blocks
-    with gr.Blocks() as blocks:
+    with gr.Blocks(css=css) as blocks:
         gr.Markdown(value=title)
 
         with gr.Tabs():
             with gr.TabItem("Upload from disk"):
+                language_radio = gr.Radio(
+                    label="Language",
+                    choices=language_choices,
+                    value=language_choices[0],
+                )
+                model_dropdown = gr.Dropdown(
+                    choices=language_to_models[language_choices[0]],
+                    label="Select a model",
+                    value=language_to_models[language_choices[0]][0],
+                )
+                decoding_method_radio = gr.Radio(
+                    label="Decoding method",
+                    choices=["greedy_search", "modified_beam_search"],
+                    value="greedy_search",
+                )
+                num_active_paths_slider = gr.Slider(
+                    minimum=1,
+                    value=4,
+                    step=1,
+                    label="Number of active paths for modified_beam_search",
+                )
+                punct_radio = gr.Radio(
+                    label="Whether to add punctuation (Only for Chinese and English)",
+                    choices=["Yes", "No"],
+                    value="Yes",
+                )
+
                 uploaded_file = gr.Audio(
                     sources=["upload"],
                     type="filepath",