import os import gradio as gr from api.audio import STTManager, TTSManager from api.llm import LLMManager from config import config from docs.instruction import instruction from resources.data import fixed_messages, topics_list from resources.prompts import prompts from utils.ui import add_candidate_message, add_interviewer_message llm = LLMManager(config, prompts) tts = TTSManager(config) stt = STTManager(config) default_audio_params = { "label": "Record answer", "sources": ["microphone"], "type": "numpy", "waveform_options": {"show_controls": False}, "editable": False, "container": False, "show_share_button": False, } def hide_settings(): init_acc = gr.Accordion("Settings", open=False) start_btn = gr.Button("Generate a problem", interactive=False) return init_acc, start_btn def show_solution(): solution_acc = gr.Accordion("Solution", open=True) end_btn = gr.Button("Finish the interview", interactive=True) audio_input = gr.Audio(interactive=True, **default_audio_params) return solution_acc, end_btn, audio_input def hide_solution(): solution_acc = gr.Accordion("Solution", open=False) end_btn = gr.Button("Finish the interview", interactive=False) problem_acc = gr.Accordion("Problem statement", open=False) audio_input = gr.Audio(interactive=False, **default_audio_params) return solution_acc, end_btn, problem_acc, audio_input def get_status_color(obj): if obj.status: if obj.streaming: return "🟢" return "🟡" return "🔴" # Interface with gr.Blocks(title="AI Interviewer") as demo: if os.getenv("IS_DEMO"): gr.Markdown(instruction["demo"]) started_coding = gr.State(False) audio_output = gr.Audio(label="Play audio", autoplay=True, visible=os.environ["DEBUG"], streaming=tts.streaming) with gr.Tab("Instruction") as instruction_tab: with gr.Row(): with gr.Column(scale=2): gr.Markdown(instruction["introduction"]) with gr.Column(scale=1): space = " " * 10 tts_status = get_status_color(tts) gr.Markdown(f"TTS status: {tts_status}{space}{config.tts.name}") try: text_test = stt.speech_to_text(audio_test, False) gr.Markdown(f"STT status: 🟢{space} {config.stt.name}") except: gr.Markdown(f"STT status: 🔴{space} {config.stt.name}") try: llm.test_connection() gr.Markdown(f"LLM status: 🟢{space} {config.llm.name}") except: gr.Markdown(f"LLM status: 🔴{space} {config.llm.name}") gr.Markdown(instruction["quick_start"]) with gr.Row(): with gr.Column(scale=2): gr.Markdown(instruction["interface"]) with gr.Column(scale=1): gr.Markdown("Bot interaction area will look like this. Use Record button to record your answer.") chat_example = gr.Chatbot( label="Chat", show_label=False, show_share_button=False, value=[["Candidate message", "Interviewer message"]] ) audio_input_example = gr.Audio(interactive=True, **default_audio_params) gr.Markdown(instruction["models"]) gr.Markdown(instruction["acknowledgements"]) with gr.Tab("Coding") as coding_tab: chat_history = gr.State([]) previous_code = gr.State("") with gr.Accordion("Settings") as init_acc: with gr.Row(): with gr.Column(): gr.Markdown("##### Problem settings") with gr.Row(): gr.Markdown("Difficulty") difficulty_select = gr.Dropdown( label="Select difficulty", choices=["Easy", "Medium", "Hard"], value="Medium", container=False, allow_custom_value=True, ) with gr.Row(): gr.Markdown("Topic (can type custom value)") topic_select = gr.Dropdown( label="Select topic", choices=topics_list, value="Arrays", container=False, allow_custom_value=True ) with gr.Column(scale=2): requirements = gr.Textbox(label="Requirements", placeholder="Specify additional requirements", lines=5) start_btn = gr.Button("Generate a problem") with gr.Accordion("Problem statement", open=True) as problem_acc: description = gr.Markdown() with gr.Accordion("Solution", open=False) as solution_acc: with gr.Row() as content: with gr.Column(scale=2): code = gr.Code( label="Please write your code here. You can use any language, but only Python syntax highlighting is available.", language="python", lines=35, ) with gr.Column(scale=1): end_btn = gr.Button("Finish the interview", interactive=False) chat = gr.Chatbot(label="Chat", show_label=False, show_share_button=False) audio_input = gr.Audio(interactive=False, **default_audio_params) message = gr.Textbox(label="Message", lines=3, visible=False) with gr.Accordion("Feedback", open=True) as feedback_acc: feedback = gr.Markdown() # Events coding_tab.select(fn=add_interviewer_message(fixed_messages["intro"]), inputs=[chat, started_coding], outputs=[chat]) start_btn.click(fn=add_interviewer_message(fixed_messages["start"]), inputs=[chat], outputs=[chat]).then( fn=lambda: True, outputs=[started_coding] ).then(fn=hide_settings, outputs=[init_acc, start_btn]).then( fn=llm.get_problem, inputs=[requirements, difficulty_select, topic_select], outputs=[description, chat_history], scroll_to_output=True, ).then( fn=show_solution, outputs=[solution_acc, end_btn, audio_input] ) end_btn.click( fn=add_interviewer_message(fixed_messages["end"]), inputs=[chat], outputs=[chat], ).then( fn=hide_solution, outputs=[solution_acc, end_btn, problem_acc, audio_input] ).then(fn=llm.end_interview, inputs=[description, chat_history], outputs=[feedback]) audio_input.stop_recording(fn=stt.speech_to_text, inputs=[audio_input], outputs=[message]).then( fn=lambda: None, outputs=[audio_input] ).then(fn=add_candidate_message, inputs=[message, chat], outputs=[chat]).then( fn=llm.send_request, inputs=[code, previous_code, message, chat_history, chat], outputs=[chat_history, chat, message, previous_code], ) chat.change(fn=tts.read_last_message, inputs=[chat], outputs=[audio_output]) demo.launch(show_api=False)