forked from SevaSk/ecoute
-
Notifications
You must be signed in to change notification settings - Fork 0
/
main.py
128 lines (95 loc) · 4.97 KB
/
main.py
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
import threading
from AudioTranscriber import AudioTranscriber
from GPTResponder import GPTResponder
import customtkinter as ctk
import AudioRecorder
import queue
import time
import torch
import sys
import TranscriberModels
import subprocess
def write_in_textbox(textbox, text):
textbox.delete("0.0", "end")
textbox.insert("0.0", text)
def update_transcript_UI(transcriber, textbox):
transcript_string = transcriber.get_transcript()
write_in_textbox(textbox, transcript_string)
textbox.after(300, update_transcript_UI, transcriber, textbox)
def update_response_UI(responder, textbox, update_interval_slider_label, update_interval_slider, freeze_state):
if not freeze_state[0]:
response = responder.response
textbox.configure(state="normal")
write_in_textbox(textbox, response)
textbox.configure(state="disabled")
update_interval = int(update_interval_slider.get())
responder.update_response_interval(update_interval)
update_interval_slider_label.configure(text=f"Update interval: {update_interval} seconds")
textbox.after(300, update_response_UI, responder, textbox, update_interval_slider_label, update_interval_slider, freeze_state)
def clear_context(transcriber, audio_queue):
transcriber.clear_transcript_data()
with audio_queue.mutex:
audio_queue.queue.clear()
def create_ui_components(root):
ctk.set_appearance_mode("dark")
ctk.set_default_color_theme("dark-blue")
root.title("Ecoute")
root.configure(bg='#252422')
root.geometry("1000x600")
font_size = 20
transcript_textbox = ctk.CTkTextbox(root, width=300, font=("Arial", font_size), text_color='#FFFCF2', wrap="word")
transcript_textbox.grid(row=0, column=0, padx=10, pady=20, sticky="nsew")
response_textbox = ctk.CTkTextbox(root, width=300, font=("Arial", font_size), text_color='#639cdc', wrap="word")
response_textbox.grid(row=0, column=1, padx=10, pady=20, sticky="nsew")
freeze_button = ctk.CTkButton(root, text="Freeze", command=None)
freeze_button.grid(row=1, column=1, padx=10, pady=3, sticky="nsew")
update_interval_slider_label = ctk.CTkLabel(root, text=f"", font=("Arial", 12), text_color="#FFFCF2")
update_interval_slider_label.grid(row=2, column=1, padx=10, pady=3, sticky="nsew")
update_interval_slider = ctk.CTkSlider(root, from_=1, to=10, width=300, height=20, number_of_steps=9)
update_interval_slider.set(2)
update_interval_slider.grid(row=3, column=1, padx=10, pady=10, sticky="nsew")
return transcript_textbox, response_textbox, update_interval_slider, update_interval_slider_label, freeze_button
def main():
try:
subprocess.run(["ffmpeg", "-version"], stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL)
except FileNotFoundError:
print("ERROR: The ffmpeg library is not installed. Please install ffmpeg and try again.")
return
root = ctk.CTk()
transcript_textbox, response_textbox, update_interval_slider, update_interval_slider_label, freeze_button = create_ui_components(root)
audio_queue = queue.Queue()
user_audio_recorder = AudioRecorder.DefaultMicRecorder()
user_audio_recorder.record_into_queue(audio_queue)
time.sleep(2)
speaker_audio_recorder = AudioRecorder.DefaultSpeakerRecorder()
speaker_audio_recorder.record_into_queue(audio_queue)
model = TranscriberModels.get_model('--api' in sys.argv)
transcriber = AudioTranscriber(user_audio_recorder.source, speaker_audio_recorder.source, model)
transcribe = threading.Thread(target=transcriber.transcribe_audio_queue, args=(audio_queue,))
transcribe.daemon = True
transcribe.start()
responder = GPTResponder()
respond = threading.Thread(target=responder.respond_to_transcriber, args=(transcriber,))
respond.daemon = True
respond.start()
print("READY")
root.grid_rowconfigure(0, weight=100)
root.grid_rowconfigure(1, weight=1)
root.grid_rowconfigure(2, weight=1)
root.grid_rowconfigure(3, weight=1)
root.grid_columnconfigure(0, weight=2)
root.grid_columnconfigure(1, weight=1)
# Add the clear transcript button to the UI
clear_transcript_button = ctk.CTkButton(root, text="Clear Transcript", command=lambda: clear_context(transcriber, audio_queue, ))
clear_transcript_button.grid(row=1, column=0, padx=10, pady=3, sticky="nsew")
freeze_state = [False] # Using list to be able to change its content inside inner functions
def freeze_unfreeze():
freeze_state[0] = not freeze_state[0] # Invert the freeze state
freeze_button.configure(text="Unfreeze" if freeze_state[0] else "Freeze")
freeze_button.configure(command=freeze_unfreeze)
update_interval_slider_label.configure(text=f"Update interval: {update_interval_slider.get()} seconds")
update_transcript_UI(transcriber, transcript_textbox)
update_response_UI(responder, response_textbox, update_interval_slider_label, update_interval_slider, freeze_state)
root.mainloop()
if __name__ == "__main__":
main()