-
Notifications
You must be signed in to change notification settings - Fork 12
/
Copy pathprocess.py
355 lines (274 loc) · 12 KB
/
process.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
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
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
try:
from ..llm import *
from .agent import *
from .chat_history import *
from ..audio.tts import text_to_speech
from ..audio.stt import speech_to_text
from ..gui.signal import signal_handler
from ..utils.db import *
from ..utils.telemetry import my_tracer, os_name
from ..utils.user_id import load_user_id
from ..version import get_version
except ImportError:
from llm import *
from agent.agent import *
from agent.chat_history import *
from audio.tts import text_to_speech
from audio.stt import speech_to_text
from gui.signal import signal_handler
from utils.db import *
from utils.user_id import load_user_id
from utils.telemetry import my_tracer, os_name
from version import get_version
import threading
import traceback
from pygame import mixer
import time
last_ai_response = None
user_id = load_user_id()
os_name_ = os_name()
def tts_if_you_can(
text: str, not_threaded=False, status_edit=False, bypass_other_settings=False
):
try:
try:
from ..gpt_computer_agent import the_main_window
except ImportError:
from gpt_computer_agent import the_main_window
first_control = None
try:
first_control = (
not is_just_text_model_active() and not the_main_window.api_enabled
)
except:
first_control = False
if first_control or bypass_other_settings:
response_path = text_to_speech(text)
if status_edit:
signal_handler.agent_response_ready.emit()
def play_audio():
for each_r in response_path:
mixer.init()
mixer.music.load(each_r)
mixer.music.play()
while mixer.music.get_busy():
the_stop_talking = False
try:
the_stop_talking = the_main_window.stop_talking
except:
pass
if the_stop_talking:
mixer.music.stop()
break
time.sleep(0.1)
if status_edit:
signal_handler.agent_response_stopped.emit()
if not not_threaded:
playback_thread = threading.Thread(target=play_audio)
playback_thread.start()
else:
play_audio()
except Exception:
traceback.print_exc()
pass
def process_audio(take_screenshot=True, take_system_audio=False, dont_save_image=False):
with my_tracer.start_span("process_audio") as span:
span.set_attribute("user_id", user_id)
span.set_attribute("os_name", os_name_)
try:
from ..audio.record import audio_data
global audio_data, last_ai_response
from ..gpt_computer_agent import the_input_box, the_main_window
from ..audio.record import audio_data
from ..audio.input_box import the_input_box_pre
transcription = speech_to_text(mic_record_location)
if take_system_audio:
transcription2 = speech_to_text(system_sound_location)
llm_input = transcription
print("Previously AI response", last_ai_response, "end prev")
print("Input Box AI", the_input_box_pre)
if (
the_input_box_pre != ""
and not the_input_box_pre.startswith("System:")
and the_input_box_pre not in last_ai_response
):
llm_input += the_input_box_pre
if take_system_audio:
llm_input += " \n Other of USER: " + transcription2
if the_input_box.toPlainText().startswith("System:"):
the_main_window.update_from_thread(
"Transciption Completed. Running AI..."
)
print("LLM INPUT (screenshot)", llm_input)
llm_output = agent(
llm_input,
get_client(),
screenshot_path=screenshot_path if take_screenshot else None,
dont_save_image=dont_save_image,
)
if the_input_box.toPlainText().startswith("System:"):
the_main_window.update_from_thread(
"AI Response Completed. Generating Audio..."
)
last_ai_response = llm_output.replace("<Answer>", "")
from ..gpt_computer_agent import the_main_window
model = load_model_settings()
if not llm_settings[model][
"stream"
] or the_main_window.worker.the_input_text.startswith("System:"):
the_main_window.set_text_to_input_box(last_ai_response)
the_main_window.complated_answer = True
signal_handler.agent_response_ready.emit()
def play_text():
from ..gpt_computer_agent import the_main_window
the_main_window.complated_answer = True
the_main_window.manuel_stop = True
while (
the_main_window.reading_thread or the_main_window.reading_thread_2
):
time.sleep(0.1)
the_main_window.read_part_task()
if the_main_window.stop_talking:
the_main_window.stop_talking = False
signal_handler.agent_response_stopped.emit()
playback_thread = threading.Thread(target=play_text)
playback_thread.start()
except Exception as e:
print("Error in process_audio", e)
traceback.print_exc()
from ..gpt_computer_agent import the_input_box, the_main_window
exception_str = traceback.format_exc()
the_main_window.update_from_thread("EXCEPTION: " + str(exception_str))
tts_if_you_can("Exception occurred. Please check the logs.")
signal_handler.agent_response_stopped.emit()
def process_screenshot():
with my_tracer.start_span("process_screenshot") as span:
span.set_attribute("user_id", user_id)
span.set_attribute("os_name", os_name_)
try:
global last_ai_response
from ..gpt_computer_agent import the_input_box, the_main_window
from ..audio.input_box import the_input_box_pre
llm_input = "I just take a screenshot. for you to remember. Just say 'Ok.' if the user doesnt want anything before."
if (
the_input_box_pre != ""
and not the_input_box_pre.startswith("System:")
and the_input_box_pre not in last_ai_response
):
llm_input += the_input_box_pre
print("LLM INPUT (just screenshot)", llm_input)
if the_input_box.toPlainText().startswith("System:"):
the_main_window.update_from_thread(
"Transciption Completed. Running AI..."
)
llm_output = agent(
llm_input,
get_client(),
screenshot_path=just_screenshot_path,
dont_save_image=False,
just_screenshot=True,
)
last_ai_response = llm_output.replace("<Answer>", "")
from ..gpt_computer_agent import the_main_window
model = load_model_settings()
if not llm_settings[model][
"stream"
] or the_main_window.worker.the_input_text.startswith("System:"):
the_main_window.set_text_to_input_box(last_ai_response)
the_main_window.complated_answer = True
signal_handler.agent_response_ready.emit()
def play_text():
from ..gpt_computer_agent import the_main_window
the_main_window.complated_answer = True
the_main_window.manuel_stop = True
while (
the_main_window.reading_thread or the_main_window.reading_thread_2
):
time.sleep(0.1)
the_main_window.read_part_task()
if the_main_window.stop_talking:
the_main_window.stop_talking = False
signal_handler.agent_response_stopped.emit()
playback_thread = threading.Thread(target=play_text)
playback_thread.start()
except Exception as e:
print("Error in process_screenshot", e)
traceback.print_exc()
from ..gpt_computer_agent import the_input_box, the_main_window
exception_str = traceback.format_exc()
the_main_window.update_from_thread("EXCEPTION: " + str(exception_str))
tts_if_you_can("Exception occurred. Please check the logs.")
signal_handler.agent_response_stopped.emit()
def process_text(text, screenshot_path=None):
with my_tracer.start_span("process_text") as span:
span.set_attribute("user_id", user_id)
span.set_attribute("os_name", os_name_)
try:
global last_ai_response
llm_input = text
llm_output = agent(
llm_input,
get_client(),
screenshot_path=screenshot_path,
dont_save_image=True,
)
last_ai_response = llm_output.replace("<Answer>", "")
from ..gpt_computer_agent import the_main_window
model = load_model_settings()
if not llm_settings[model][
"stream"
] or the_main_window.worker.the_input_text.startswith("System:"):
the_main_window.set_text_to_input_box(last_ai_response)
the_main_window.complated_answer = True
signal_handler.agent_response_ready.emit()
def play_text():
from ..gpt_computer_agent import the_main_window
the_main_window.complated_answer = True
the_main_window.manuel_stop = True
while (
the_main_window.reading_thread or the_main_window.reading_thread_2
):
time.sleep(0.1)
the_main_window.read_part_task()
if the_main_window.stop_talking:
the_main_window.stop_talking = False
signal_handler.agent_response_stopped.emit()
playback_thread = threading.Thread(target=play_text)
playback_thread.start()
except Exception as e:
print("Error in process_text", e)
traceback.print_exc()
from ..gpt_computer_agent import the_main_window
exception_str = traceback.format_exc()
the_main_window.update_from_thread("EXCEPTION: " + str(exception_str))
tts_if_you_can("Exception occurred. Please check the logs.")
signal_handler.agent_response_stopped.emit()
import sentry_sdk
sentry_sdk.init(
dsn="https://eed76b3c8eb23bbe1c2f6a796a03f1a9@o4508336623583232.ingest.us.sentry.io/4508556319195136",
# Set traces_sample_rate to 1.0 to capture 100%
# of transactions for tracing.
traces_sample_rate=1.0,
release=f"gcs@{get_version()}",
server_name="gca_client",
)
def process_text_api(text, screenshot_path=None):
with my_tracer.start_span("process_text_api") as span:
span.set_attribute("user_id", user_id)
span.set_attribute("os_name", os_name_)
try:
global last_ai_response
llm_input = text
sentry_sdk.set_user({"id": load_user_id()})
sentry_sdk.profiler.start_profiler()
llm_output = agent(
llm_input,
get_client(),
screenshot_path=screenshot_path,
dont_save_image=True,
)
sentry_sdk.profiler.stop_profiler()
return llm_output
except Exception as e:
print("Error in process_text", e)
traceback.print_exc()