Update app.py

app.py

@@ -339,6 +339,7 @@ def generate_subtitles(audio_file_path, prompt, language, auto_detect_language,
     return temp_srt_path, None, None 
     
 
+
 with gr.Blocks() as demo:
     gr.Markdown(
         """
@@ -349,65 +350,63 @@ with gr.Blocks() as demo:
     )
     with gr.Tabs():
         with gr.TabItem("LLMs"):
-            with gr.Row():
-                with gr.Column():
-                    model = gr.Dropdown(
-                        choices=[
-                            "llama3-70b-8192",
-                            "llama3-8b-8192",
-                            "mixtral-8x7b-32768",
-                            "gemma-7b-it",
-                            "gemma2-9b-it",
-                        ],
-                        value="llama3-70b-8192",
-                        label="Model",
-                    )
-                    temperature = gr.Slider(
-                        minimum=0.0,
-                        maximum=1.0,
-                        step=0.01,
-                        value=0.5,
-                        label="Temperature",
-                        info="Controls diversity of the generated text. Lower is more deterministic, higher is more creative.",
-                    )
-                    max_tokens = gr.Slider(
-                        minimum=1,
-                        maximum=8192,
-                        step=1,
-                        value=4096,
-                        label="Max Tokens",
-                        info="The maximum number of tokens that the model can process in a single response.<br>Maximums: 8k for gemma 7b it, gemma2 9b it, llama 7b & 70b, 32k for mixtral 8x7b.",
-                    )
-                    top_p = gr.Slider(
-                        minimum=0.0,
-                        maximum=1.0,
-                        step=0.01,
-                        value=0.5,
-                        label="Top P",
-                        info="A method of text generation where a model will only consider the most probable next tokens that make up the probability p.",
-                    )
-                    seed = gr.Number(
-                        precision=0, value=42, label="Seed", info="A starting point to initiate generation, use 0 for random"
-                    )
-                    model.change(update_max_tokens, inputs=[model], outputs=max_tokens)
-                with gr.Column():
-                    chatbot = gr.ChatInterface(
-                        fn=generate_response,
-                        chatbot=None,
-                        additional_inputs=[
-                            model,
-                            temperature,
-                            max_tokens,
-                            top_p,
-                            seed,
-                        ],
-                    )
-                    model.change(update_max_tokens, inputs=[model], outputs=max_tokens)
+            with gr.Column():
+                model = gr.Dropdown(
+                    choices=[
+                        "llama3-70b-8192",
+                        "llama3-8b-8192",
+                        "mixtral-8x7b-32768",
+                        "gemma-7b-it",
+                        "gemma2-9b-it",
+                    ],
+                    value="llama3-70b-8192",
+                    label="Model",
+                )
+                temperature = gr.Slider(
+                    minimum=0.0,
+                    maximum=1.0,
+                    step=0.01,
+                    value=0.5,
+                    label="Temperature",
+                    info="Controls diversity of the generated text. Lower is more deterministic, higher is more creative.",
+                )
+                max_tokens = gr.Slider(
+                    minimum=1,
+                    maximum=8192,
+                    step=1,
+                    value=4096,
+                    label="Max Tokens",
+                    info="The maximum number of tokens that the model can process in a single response.<br>Maximums: 8k for gemma 7b it, gemma2 9b it, llama 7b & 70b, 32k for mixtral 8x7b.",
+                )
+                top_p = gr.Slider(
+                    minimum=0.0,
+                    maximum=1.0,
+                    step=0.01,
+                    value=0.5,
+                    label="Top P",
+                    info="A method of text generation where a model will only consider the most probable next tokens that make up the probability p.",
+                )
+                seed = gr.Number(
+                    precision=0, value=42, label="Seed", info="A starting point to initiate generation, use 0 for random"
+                )
+                model.change(update_max_tokens, inputs=[model], outputs=max_tokens)
+                chatbot = gr.ChatInterface(
+                    fn=generate_response,
+                    chatbot=None,
+                    additional_inputs=[
+                        model,
+                        temperature,
+                        max_tokens,
+                        top_p,
+                        seed,
+                    ],
+                )
+                model.change(update_max_tokens, inputs=[model], outputs=max_tokens)
         with gr.TabItem("Speech To Text"):
             with gr.Tabs():
                 with gr.TabItem("Transcription"):
                     gr.Markdown("Transcript audio from files to text!")
-                    with gr.Row():
+                    with gr.Column():
                         audio_input = gr.File(
                             type="filepath", label="Upload File containing Audio", file_types=[f".{ext}" for ext in ALLOWED_FILE_EXTENSIONS]
                         )
@@ -416,12 +415,10 @@ with gr.Blocks() as demo:
                             value="whisper-large-v3",
                             label="Model",
                         )
-                    with gr.Row():
                         transcribe_prompt = gr.Textbox(
                             label="Prompt (Optional)",
                             info="Specify any context or spelling corrections.",
                         )
-                    with gr.Column():
                         language = gr.Dropdown(
                             choices=[(lang, code) for lang, code in LANGUAGE_CODES.items()],
                             value="en",
@@ -437,7 +434,7 @@ with gr.Blocks() as demo:
                     )
                 with gr.TabItem("Translation"):
                     gr.Markdown("Transcript audio from files and translate them to English text!")
-                    with gr.Row():
+                    with gr.Column():
                         audio_input_translate = gr.File(
                             type="filepath", label="Upload File containing Audio", file_types=[f".{ext}" for ext in ALLOWED_FILE_EXTENSIONS]
                         )
@@ -446,7 +443,6 @@ with gr.Blocks() as demo:
                             value="whisper-large-v3",
                             label="Model",
                         )
-                    with gr.Row():
                         translate_prompt = gr.Textbox(
                             label="Prompt (Optional)",
                             info="Specify any context or spelling corrections.",
@@ -459,7 +455,7 @@ with gr.Blocks() as demo:
                         outputs=translation_output,
                     )
                 with gr.TabItem("Subtitle Maker"):
-                    with gr.Row():
+                    with gr.Column():
                         audio_input_subtitles = gr.File(
                             label="Upload Audio/Video",
                             file_types=[f".{ext}" for ext in ALLOWED_FILE_EXTENSIONS],
@@ -473,15 +469,14 @@ with gr.Blocks() as demo:
                             label="Prompt (Optional)",
                             info="Specify any context or spelling corrections.",
                         )
-                        with gr.Row():
-                            language_subtitles = gr.Dropdown(
-                                choices=[(lang, code) for lang, code in LANGUAGE_CODES.items()],
-                                value="en",
-                                label="Language",
-                            )
-                            auto_detect_language_subtitles = gr.Checkbox(
-                                label="Auto Detect Language"
-                            )
+                        language_subtitles = gr.Dropdown(
+                            choices=[(lang, code) for lang, code in LANGUAGE_CODES.items()],
+                            value="en",
+                            label="Language",
+                        )
+                        auto_detect_language_subtitles = gr.Checkbox(
+                            label="Auto Detect Language"
+                        )
                     transcribe_button_subtitles = gr.Button("Generate Subtitles")
                     srt_output = gr.File(label="SRT Output File")
                     video_output = gr.File(label="Output Video with Subtitles")
@@ -497,4 +492,5 @@ with gr.Blocks() as demo:
                         outputs=[srt_output, video_output, gr.Textbox(label="Error")],
                     )
                     
-demo.launch()
+demo.launch()
+