File size: 7,868 Bytes
523a4a0
 
 
 
6e988a5
523a4a0
 
 
 
 
 
 
 
 
 
54f604e
523a4a0
 
 
6e988a5
523a4a0
 
 
2b106d0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
083e171
523a4a0
083e171
 
 
6e988a5
 
 
 
 
 
 
 
 
 
083e171
523a4a0
 
 
 
 
 
b78042d
 
523a4a0
 
 
 
14d2111
 
083e171
 
14d2111
2b88d53
14d2111
 
2b106d0
 
083e171
54f604e
 
8e1a910
083e171
 
 
54f604e
 
 
 
 
 
 
083e171
 
6e988a5
083e171
 
 
 
 
6e988a5
083e171
 
 
 
 
6e988a5
 
 
 
 
 
 
 
083e171
 
 
 
 
6e988a5
 
 
083e171
6e988a5
 
 
 
 
083e171
6e988a5
 
083e171
 
 
 
6e988a5
083e171
6e988a5
083e171
54f604e
083e171
 
 
54f604e
083e171
523a4a0
 
2b106d0
 
 
54f604e
2b106d0
 
523a4a0
 
083e171
 
523a4a0
083e171
523a4a0
 
 
 
 
 
 
 
 
 
 
 
 
 
083e171
523a4a0
 
6e988a5
 
 
 
 
523a4a0
 
 
 
083e171
523a4a0
 
 
 
 
 
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
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
import gradio as gr
import tempfile
import os
from pathlib import Path
from io import BytesIO
from settings import (
    respond,
    generate_random_string,
    reset_interview,
    generate_interview_report,
    generate_report_from_file,
    interview_history,
    question_count,
    language,
)
from ai_config import convert_text_to_speech, transcribe_audio, n_of_questions
from prompt_instructions import get_interview_initial_message

# Global variables
temp_audio_files = []
initial_audio_path = None


def reset_interview_action():
    global question_count, interview_history
    question_count = 0
    interview_history.clear()
    initial_message = get_interview_initial_message()

    # Generate new audio for the initial message
    initial_audio_buffer = BytesIO()
    convert_text_to_speech(initial_message, initial_audio_buffer)
    initial_audio_buffer.seek(0)

    with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as temp_file:
        temp_audio_path = temp_file.name
        temp_file.write(initial_audio_buffer.getvalue())

    temp_audio_files.append(temp_audio_path)

    return (
        [(None, initial_message)],  # Reset chatbot
        temp_audio_path,  # New audio
        gr.File(visible=False),  # Hide PDF output
        gr.Textbox(visible=True),  # Show message input
        "Interview reset. You can start a new interview now."  # Status message
    )

# Initialize Gradio interface
def create_app():
    global initial_audio_path
    initial_message = get_interview_initial_message()

    # Generate the audio for the initial message and save to a temporary file
    initial_audio_buffer = BytesIO()
    convert_text_to_speech(initial_message, initial_audio_buffer)
    initial_audio_buffer.seek(0)

    with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as temp_file:
        initial_audio_path = temp_file.name
        temp_file.write(initial_audio_buffer.getvalue())

    temp_audio_files.append(initial_audio_path)

    with gr.Blocks(title="Clinical Psychologist Interviewer ๐šฟ") as demo:
        gr.Markdown(
            """
            # Clinical Psychologist Interviewer ๐šฟ
            This chatbot conducts clinical interviews based on psychological knowledge.
            Please note that this is a simulation and should not be used as a substitute for professional medical advice.
            
            The interviewer will prepare a clinical report based on the interview.
            """
        )

        with gr.Tab("Interview"):
            audio_output = gr.Audio(
                label="Sarah",
                scale=1,
                value=initial_audio_path,
                autoplay=True,
                visible=True,
                show_download_button=False,
            )

            reset_button = gr.Button("Reset Interview", size='sm')
            chatbot = gr.Chatbot(value=[(None, f"{initial_message}")], label=f"Clinical Interview ๐šฟ๐Ÿ“‹")
            with gr.Row():
                msg = gr.Textbox(label="Type your message here...", scale=3)
                audio_input = gr.Audio(sources=(["microphone"]), label="Record your message", type="filepath", scale=2)
            send_button = gr.Button("Send")
            pdf_output = gr.File(label="Download Report", visible=False)

            def user(user_message, audio, history):
                print(audio)
                if audio is not None:
                    user_message = transcribe_audio(audio)
                    print(user_message)

                return "", None, history + [[user_message, None]]

            def bot_response(chatbot, message):
                global question_count, temp_audio_files
                question_count += 1

                # Use the last user message from the chatbot history
                last_user_message = chatbot[-1][0] if chatbot else message

                response, audio_buffer = respond(chatbot, last_user_message)

                # Add all bot responses to the chatbot history
                for bot_message in response:
                    chatbot.append((None, bot_message[1]))

                if isinstance(audio_buffer, BytesIO):
                    with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as temp_file:
                        temp_audio_path = temp_file.name
                        temp_file.write(audio_buffer.getvalue())
                    temp_audio_files.append(temp_audio_path)
                    audio_output = temp_audio_path
                else:
                    audio_output = audio_buffer

                if question_count >= n_of_questions():
                    conclusion_message = "Thank you for participating in this interview. We have reached the end of our session. I hope this conversation has been helpful. Take care!"
                    chatbot.append((None, conclusion_message))

                    conclusion_audio_buffer = BytesIO()
                    convert_text_to_speech(conclusion_message, conclusion_audio_buffer)
                    conclusion_audio_buffer.seek(0)

                    with tempfile.NamedTemporaryFile(suffix=".mp3", delete=False) as temp_file:
                        temp_audio_path = temp_file.name
                        temp_file.write(conclusion_audio_buffer.getvalue())
                    temp_audio_files.append(temp_audio_path)
                    audio_output = temp_audio_path

                    # Generate report automatically
                    report_content, pdf_path = generate_interview_report(interview_history, language)

                    # Add report to the chat
                    chatbot.append((None, f"Interview Report:\n\n{report_content}"))

                    return chatbot, audio_output, gr.File(visible=True, value=pdf_path), gr.Textbox(visible=False)

                return chatbot, audio_output, gr.File(visible=False), gr.Textbox(visible=True)

            msg.submit(user, [msg, audio_input, chatbot], [msg, audio_input, chatbot], queue=False).then(
                bot_response, [chatbot, msg], [chatbot, audio_output, pdf_output, msg]
            )

            send_button.click(user, [msg, audio_input, chatbot], [msg, audio_input, chatbot], queue=False).then(
                bot_response, [chatbot, msg], [chatbot, audio_output, pdf_output, msg]
            )

            reset_button.click(
                reset_interview_action,
                inputs=[],
                outputs=[chatbot, audio_output, pdf_output, msg, audio_input]
            )

        with gr.Tab("Upload Document"):
            file_input = gr.File(label="Upload a TXT, PDF, or DOCX file")
            language_input = gr.Textbox(label="Preferred Language for Report",
                                        placeholder="Enter language")
            generate_button = gr.Button("Generate Report")
            report_output = gr.Textbox(label="Generated Report", lines=100)
            pdf_output = gr.File(label="Download Report", visible=True)

            def generate_report_and_pdf(file, language):
                report_content, pdf_path = generate_report_from_file(file, language)
                return report_content, pdf_path, gr.File(visible=True)

            generate_button.click(
                generate_report_and_pdf,
                inputs=[file_input, language_input],
                outputs=[report_output, pdf_output, pdf_output]
            )

    return demo


# Clean up function
def cleanup():
    global temp_audio_files, initial_audio_path
    for audio_file in temp_audio_files:
        if os.path.exists(audio_file):
            os.unlink(audio_file)
    temp_audio_files.clear()

    if initial_audio_path and os.path.exists(initial_audio_path):
        os.unlink(initial_audio_path)


if __name__ == "__main__":
    app = create_app()
    try:
        app.launch()
    finally:
        cleanup()