app.py

"""import gradio as gr
from huggingface_hub import InferenceClient


def respond(
    message,
    history: list[dict[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
    hf_token: gr.OAuthToken,
):
    
    #For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
    
    client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")

    messages = [{"role": "system", "content": system_message}]

    messages.extend(history)

    messages.append({"role": "user", "content": message})

    response = ""

    for message in client.chat_completion(
        messages,
        max_tokens=max_tokens,
        stream=True,
        temperature=temperature,
        top_p=top_p,
    ):
        choices = message.choices
        token = ""
        if len(choices) and choices[0].delta.content:
            token = choices[0].delta.content

        response += token
        yield response



#For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface

chatbot = gr.ChatInterface(
    respond,
    type="messages",
    additional_inputs=[
        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.95,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
    ],
)

with gr.Blocks() as demo:
    with gr.Sidebar():
        gr.LoginButton()
    chatbot.render()


if __name__ == "__main__":
    demo.launch()
"""
import gradio as gr
import requests
from huggingface_hub import InferenceClient

DEEPGRAM_API_KEY = "0c72698eb40f85fc25b56a76039e795be653afed"

def deepgram_stt(audio_file_path):
    
    #Send user microphone audio to Deepgram STT
    
    url = "https://api.deepgram.com/v1/listen"
    headers = {
        "Authorization": f"Token {DEEPGRAM_API_KEY}",
        "Content-Type": "audio/wav"
    }

    with open(audio_file_path, "rb") as f:
        audio = f.read()

    response = requests.post(url, headers=headers, data=audio).json()
    return response["results"]["channels"][0]["alternatives"][0]["transcript"]


def deepgram_tts(text):
    
    #Convert model output → speech using Deepgram TTS
    
    url = "https://api.deepgram.com/v1/speak?model=aura-asteria-en"  # any model
    headers = {
        "Authorization": f"Token {DEEPGRAM_API_KEY}",
        "Content-Type": "application/json"
    }

    payload = {"text": text}

    audio_out = "response.wav"
    r = requests.post(url, json=payload, headers=headers)

    with open(audio_out, "wb") as f:
        f.write(r.content)

    return audio_out


def respond_audio(
    audio_input,
    history,
    system_message,
    max_tokens,
    temperature,
    top_p,
    hf_token: gr.OAuthToken,
):
    
    #STT → send to model → TTS
    
    client = InferenceClient(
        token=hf_token.token,
        model="openai/gpt-oss-20b"
    )

    # ---- 1. Speech → text ----
    user_message = deepgram_stt(audio_input)

    messages = [{"role": "system", "content": system_message}]
    messages.extend(history)
    messages.append({"role": "user", "content": user_message})

    # ---- 2. Model response ----
    response_text = ""
    for message in client.chat_completion(
        messages,
        max_tokens=max_tokens,
        stream=True,
        temperature=temperature,
        top_p=top_p,
    ):
        if len(message.choices) and message.choices[0].delta.content:
            response_text += message.choices[0].delta.content
            yield response_text, None  # update text while streaming

    # ---- 3. Text → audio ----
    audio_file = deepgram_tts(response_text)

    yield response_text, audio_file


with gr.Blocks() as demo:
    with gr.Sidebar():
        gr.LoginButton()

    gr.Markdown("## 🎤 Voice Chat Mode (Deepgram + GPT-OSS)")

    # Hidden but expandable textbox
    with gr.Accordion("Optional: Type Instead of Speaking", open=False):
        typed_message = gr.Textbox(label="Manual Text Input")
    
    chatbot = gr.Chatbot(type="messages")
    
    audio_in = gr.Audio(label="Press to Speak", type="filepath")
    audio_out = gr.Audio(label="TTS Output")

    system_message = gr.Textbox(
        value="You are a friendly Chatbot.",
        label="System message"
    )
    max_tokens = gr.Slider(1, 2048, value=512, label="Max new tokens")
    temp = gr.Slider(0.1, 4.0, value=0.7, label="Temperature")
    top_p = gr.Slider(0.1, 1.0, value=0.95, label="Top-p")

    send_button = gr.Button("Send (Voice)")

    send_button.click(
        respond_audio,
        inputs=[audio_in, chatbot, system_message, max_tokens, temp, top_p],
        outputs=[chatbot, audio_out]
    )

if __name__ == "__main__":
    demo.launch()