Update app.py

app.py

@@ -340,6 +340,7 @@ def generate_subtitles(audio_file_path, prompt, language, auto_detect_language,
     
 
 
+
 with gr.Blocks() as demo:
     gr.Markdown(
         """
@@ -350,148 +351,148 @@ with gr.Blocks() as demo:
     )
     with gr.Tabs():
         with gr.TabItem("select option here:"):
-                with gr.Tabs():
-                    with gr.TabItem("Speech To Text"):
-                        
-                    with gr.TabItem("Transcription"):
-                        gr.Markdown("Transcript audio from files to text!")
-                        with gr.Column():
-                            audio_input = gr.File(
-                                type="filepath", label="Upload File containing Audio", file_types=[f".{ext}" for ext in ALLOWED_FILE_EXTENSIONS]
-                            )
-                            model_choice_transcribe = gr.Dropdown(
-                                choices=["whisper-large-v3"],  # Only include 'whisper-large-v3'
-                                value="whisper-large-v3",
-                                label="Model",
-                            )
-                            transcribe_prompt = gr.Textbox(
-                                label="Prompt (Optional)",
-                                info="Specify any context or spelling corrections.",
-                            )
-                            language = gr.Dropdown(
-                                 choices=[(lang, code) for lang, code in LANGUAGE_CODES.items()],
-                                 value="en",
-                                 label="Language",
-                            )
-                            auto_detect_language = gr.Checkbox(label="Auto Detect Language")
-                            transcribe_button = gr.Button("Transcribe")
-                        transcription_output = gr.Textbox(label="Transcription")
-                        transcribe_button.click(
-                            transcribe_audio,
-                            inputs=[audio_input, transcribe_prompt, language, auto_detect_language, model_choice_transcribe],
-                            outputs=transcription_output,
+            with gr.Tabs():
+                with gr.TabItem("Speech To Text"):
+                    gr.Markdown("Speech to Text coming soon!")
+                with gr.TabItem("Transcription"):
+                    gr.Markdown("Transcript audio from files to text!")
+                    with gr.Column():
+                        audio_input = gr.File(
+                            type="filepath", label="Upload File containing Audio", file_types=[f".{ext}" for ext in ALLOWED_FILE_EXTENSIONS]
                         )
-            with gr.TabItem("Translation"):
-                gr.Markdown("Transcript audio from files and translate them to English text!")
-                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]
-                    )
-                    model_choice_translate = gr.Dropdown(
-                        choices=["whisper-large-v3"],  # Only include 'whisper-large-v3'
-                        value="whisper-large-v3",
-                        label="Model",
-                    )
-                    translate_prompt = gr.Textbox(
-                        label="Prompt (Optional)",
-                        info="Specify any context or spelling corrections.",
+                        model_choice_transcribe = gr.Dropdown(
+                            choices=["whisper-large-v3"],  # Only include 'whisper-large-v3'
+                            value="whisper-large-v3",
+                            label="Model",
+                        )
+                        transcribe_prompt = gr.Textbox(
+                            label="Prompt (Optional)",
+                            info="Specify any context or spelling corrections.",
+                        )
+                        language = gr.Dropdown(
+                             choices=[(lang, code) for lang, code in LANGUAGE_CODES.items()],
+                             value="en",
+                             label="Language",
+                        )
+                        auto_detect_language = gr.Checkbox(label="Auto Detect Language")
+                        transcribe_button = gr.Button("Transcribe")
+                    transcription_output = gr.Textbox(label="Transcription")
+                    transcribe_button.click(
+                        transcribe_audio,
+                        inputs=[audio_input, transcribe_prompt, language, auto_detect_language, model_choice_transcribe],
+                        outputs=transcription_output,
                     )
-                    translate_button = gr.Button("Translate")
+                with gr.TabItem("Translation"):
+                    gr.Markdown("Transcript audio from files and translate them to English text!")
+                    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]
+                        )
+                        model_choice_translate = gr.Dropdown(
+                            choices=["whisper-large-v3"],  # Only include 'whisper-large-v3'
+                            value="whisper-large-v3",
+                            label="Model",
+                        )
+                        translate_prompt = gr.Textbox(
+                            label="Prompt (Optional)",
+                            info="Specify any context or spelling corrections.",
+                        )
+                        translate_button = gr.Button("Translate")
                     translation_output = gr.Textbox(label="Translation")
                     translate_button.click(
                         translate_audio,
                         inputs=[audio_input_translate, translate_prompt, model_choice_translate],
                         outputs=translation_output,
                     )
-            with gr.TabItem("Subtitle Maker"):
-                with gr.Column():
-                    audio_input_subtitles = gr.File(
-                        label="Upload Audio/Video",
-                        file_types=[f".{ext}" for ext in ALLOWED_FILE_EXTENSIONS],
-                    )
-                    model_choice_subtitles = gr.Dropdown(
-                        choices=["whisper-large-v3"],  # Only include 'whisper-large-v3'
-                        value="whisper-large-v3",
-                        label="Model",
-                    )
-                    transcribe_prompt_subtitles = gr.Textbox(
-                        label="Prompt (Optional)",
-                        info="Specify any context or spelling corrections.",
-                    )
-                    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")
-                    transcribe_button_subtitles.click(
-                        generate_subtitles,
-                        inputs=[
-                            audio_input_subtitles,
-                            transcribe_prompt_subtitles,
-                            language_subtitles,
-                            auto_detect_language_subtitles,
-                            model_choice_subtitles,
-                        ],
-                        outputs=[srt_output, video_output, gr.Textbox(label="Error")]
-                    )
-            with gr.TabItem("LLMs"):
-                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,
+                with gr.TabItem("Subtitle Maker"):
+                    with gr.Column():
+                        audio_input_subtitles = gr.File(
+                            label="Upload Audio/Video",
+                            file_types=[f".{ext}" for ext in ALLOWED_FILE_EXTENSIONS],
+                        )
+                        model_choice_subtitles = gr.Dropdown(
+                            choices=["whisper-large-v3"],  # Only include 'whisper-large-v3'
+                            value="whisper-large-v3",
+                            label="Model",
+                        )
+                        transcribe_prompt_subtitles = gr.Textbox(
+                            label="Prompt (Optional)",
+                            info="Specify any context or spelling corrections.",
+                        )
+                        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")
+                        transcribe_button_subtitles.click(
+                            generate_subtitles,
+                            inputs=[
+                                audio_input_subtitles,
+                                transcribe_prompt_subtitles,
+                                language_subtitles,
+                                auto_detect_language_subtitles,
+                                model_choice_subtitles,
+                            ],
+                            outputs=[srt_output, video_output, gr.Textbox(label="Error")]
+                        )
+                with gr.TabItem("LLMs"):
+                    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)
+                            ],
+                        )
+                        model.change(update_max_tokens, inputs=[model], outputs=max_tokens)
 
 demo.launch()