File size: 3,469 Bytes
e6868fd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
import gradio as gr
from transformers import pipeline
import numpy as np
from openai import OpenAI

transcriber = pipeline("automatic-speech-recognition", model="openai/whisper-base.en")
qa_model = pipeline("question-answering", model="distilbert-base-cased-distilled-squad")

def predict(message, history, api_key):
    print('in predict')
    client = OpenAI(api_key=api_key)
    history_openai_format = []
    for human, assistant in history:
        history_openai_format.append({"role": "user", "content": human})
        history_openai_format.append({"role": "assistant", "content": assistant})
    history_openai_format.append({"role": "user", "content": message})

    response = client.chat.completions.create(
        model='gpt-4o',
        messages=history_openai_format,
        temperature=1.0,
        stream=True
    )

    partial_message = ""
    for chunk in response:
        if chunk.choices[0].delta.content:
            print(111, chunk.choices[0].delta.content)
            partial_message += chunk.choices[0].delta.content
            yield partial_message

def chat_with_api_key(api_key, message, history):
    print('in chat_with_api_key')
    accumulated_message = ""
    for partial_message in predict(message, history, api_key):
        accumulated_message = partial_message
        history.append((message, accumulated_message))
        # yield accumulated_message, history
        yield message,[[message, accumulated_message]]

def transcribe(audio):
    if audio is None:
        return "No audio recorded."
    sr, y = audio
    y = y.astype(np.float32)
    y /= np.max(np.abs(y))
    
    return transcriber({"sampling_rate": sr, "raw": y})["text"]

def answer(transcription):
    context = "You are a chatbot answering general questions"
    result = qa_model(question=transcription, context=context)
    return result['answer']

def process_audio(audio):
    if audio is None:
        return "No audio recorded.", []
    transcription = transcribe(audio)
    answer_result = answer(transcription)
    return transcription, [[transcription, answer_result]]

def update_output(api_key, audio_input, state):
    print('in update_output')
    message = transcribe(audio_input)
    responses = chat_with_api_key(api_key, message, state)
    accumulated_response = ""
    for response, updated_state in responses:
        accumulated_response = response
        yield accumulated_response, updated_state

def clear_all():
    return None, "", []

with gr.Blocks() as demo:
    answer_output = gr.Chatbot(label="Answer Result")
    with gr.Row():    
        audio_input = gr.Audio(label="Audio Input", sources=["microphone"], type="numpy")
        with gr.Column():
            api_key = gr.Textbox(label="API Key", placeholder="Enter your API key", type="password")
            transcription_output = gr.Textbox(label="Transcription")
            clear_button = gr.Button("Clear")
    state = gr.State([])
    if 1:
        audio_input.stop_recording(
            fn=update_output,
            inputs=[api_key, audio_input, state],
            outputs=[transcription_output, answer_output]
        )
    if 0:
        audio_input.stop_recording(
            fn=process_audio,
            inputs=[audio_input],
            outputs=[transcription_output, answer_output]
        )
    
    clear_button.click(
        fn=clear_all,
        inputs=[],
        outputs=[audio_input, transcription_output, answer_output]
    )

demo.launch()