Update app.py
Browse files
app.py
CHANGED
|
@@ -10,8 +10,19 @@ tokenizer = GPT2Tokenizer.from_pretrained('gpt2')
|
|
| 10 |
import whisper
|
| 11 |
|
| 12 |
import os
|
| 13 |
-
import dropbox
|
| 14 |
import datetime
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
|
| 16 |
|
| 17 |
ACCESS_TOKEN = os.environ["ACCESS_TOKEN"]
|
|
@@ -85,7 +96,14 @@ def transcribe(audio, text):
|
|
| 85 |
|
| 86 |
# Add the input text to the messages list
|
| 87 |
messages.append({"role": "user", "content": input_text})
|
| 88 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 89 |
# Check if the accumulated tokens have exceeded 2096
|
| 90 |
num_tokens = num_tokens_from_messages(messages)
|
| 91 |
if num_tokens > 2096:
|
|
@@ -98,9 +116,19 @@ def transcribe(audio, text):
|
|
| 98 |
|
| 99 |
chat_transcript_copy = chat_transcript
|
| 100 |
chat_transcript_copy += f"Number of tokens used: {num_tokens}\n\n"
|
| 101 |
-
|
| 102 |
-
|
| 103 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 104 |
|
| 105 |
if num_tokens > 2200:
|
| 106 |
# Reset the messages list and answer counter
|
|
@@ -128,14 +156,27 @@ def transcribe(audio, text):
|
|
| 128 |
chat_transcript += f"[ANSWER {answer_count}]" + message['role'] + ": " + message['content'] + "\n\n"
|
| 129 |
# Append the number of tokens used to the end of the chat transcript
|
| 130 |
|
| 131 |
-
with open("
|
| 132 |
f.write(chat_transcript)
|
| 133 |
|
| 134 |
chat_transcript_copy = chat_transcript
|
| 135 |
chat_transcript_copy += f"Number of tokens used: {num_tokens}\n\n"
|
| 136 |
-
filename = datetime.datetime.now().strftime("%
|
| 137 |
-
|
| 138 |
-
dbx.files_upload(chat_transcript_copy.encode('utf-8'), '/
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 139 |
|
| 140 |
return chat_transcript
|
| 141 |
|
|
@@ -163,8 +204,6 @@ iface.launch()
|
|
| 163 |
|
| 164 |
|
| 165 |
|
| 166 |
-
|
| 167 |
-
|
| 168 |
# from transformers import pipeline, T5Tokenizer
|
| 169 |
# import pyttsx3
|
| 170 |
# import threading
|
|
|
|
| 10 |
import whisper
|
| 11 |
|
| 12 |
import os
|
|
|
|
| 13 |
import datetime
|
| 14 |
+
# import dropbox
|
| 15 |
+
# from notion_client import Client
|
| 16 |
+
import notion_df
|
| 17 |
+
|
| 18 |
+
|
| 19 |
+
# # Define your API key
|
| 20 |
+
|
| 21 |
+
# my_API_KEY = os.environ["NOTION"]
|
| 22 |
+
# notion = Client(auth=my_API_KEY)
|
| 23 |
+
# # find the page you want to upload the file to
|
| 24 |
+
# page = notion.pages.retrieve(page_id="37660063895a4525b5cd8feffd43f5d5")
|
| 25 |
+
|
| 26 |
|
| 27 |
|
| 28 |
ACCESS_TOKEN = os.environ["ACCESS_TOKEN"]
|
|
|
|
| 96 |
|
| 97 |
# Add the input text to the messages list
|
| 98 |
messages.append({"role": "user", "content": input_text})
|
| 99 |
+
|
| 100 |
+
# Get the current date and time in the local timezone
|
| 101 |
+
now_local = datetime.datetime.now()
|
| 102 |
+
# Create a timezone object for Eastern Time (ET)
|
| 103 |
+
et_tz = datetime.timezone(datetime.timedelta(hours=-5))
|
| 104 |
+
# Adjust the date and time to Eastern Time (ET)
|
| 105 |
+
now_et = now_local.astimezone(et_tz)
|
| 106 |
+
|
| 107 |
# Check if the accumulated tokens have exceeded 2096
|
| 108 |
num_tokens = num_tokens_from_messages(messages)
|
| 109 |
if num_tokens > 2096:
|
|
|
|
| 116 |
|
| 117 |
chat_transcript_copy = chat_transcript
|
| 118 |
chat_transcript_copy += f"Number of tokens used: {num_tokens}\n\n"
|
| 119 |
+
|
| 120 |
+
# Get the current UTC time
|
| 121 |
+
utc_time = datetime.now(timezone.utc)
|
| 122 |
+
# Convert to Eastern Time Zone
|
| 123 |
+
eastern_time = utc_time + timedelta(hours=-5)
|
| 124 |
+
# Format as string (YY-MM-DD HH:MM)
|
| 125 |
+
published_date = eastern_time.strftime('%m-%d-%y %H:%M')
|
| 126 |
+
|
| 127 |
+
import pandas as pd
|
| 128 |
+
# string dataframe?
|
| 129 |
+
df = pd.DataFrame([chat_transcript])
|
| 130 |
+
notion_df.upload(df, 'https://www.notion.so/page-827360c361f347f7bfefcc6dfbd10e51', title=str(published_date), api_key=API_KEY)
|
| 131 |
+
|
| 132 |
|
| 133 |
if num_tokens > 2200:
|
| 134 |
# Reset the messages list and answer counter
|
|
|
|
| 156 |
chat_transcript += f"[ANSWER {answer_count}]" + message['role'] + ": " + message['content'] + "\n\n"
|
| 157 |
# Append the number of tokens used to the end of the chat transcript
|
| 158 |
|
| 159 |
+
with open("conversation_history.txt", "a") as f:
|
| 160 |
f.write(chat_transcript)
|
| 161 |
|
| 162 |
chat_transcript_copy = chat_transcript
|
| 163 |
chat_transcript_copy += f"Number of tokens used: {num_tokens}\n\n"
|
| 164 |
+
filename = datetime.datetime.now().strftime("%m%d%y_%H:%M_conversation_history.txt")
|
| 165 |
+
|
| 166 |
+
# dbx.files_upload(chat_transcript_copy.encode('utf-8'), f'/{filename}', mode=dropbox.files.WriteMode.overwrite, autorename=False, client_modified=None, mute=False)
|
| 167 |
+
# dbx.files_upload(chat_transcript_copy.encode('utf-8'), '/conversation_history.txt', mode=dropbox.files.WriteMode.overwrite, autorename=False, client_modified=None, mute=False)
|
| 168 |
+
|
| 169 |
+
# Get the current UTC time
|
| 170 |
+
utc_time = datetime.now(timezone.utc)
|
| 171 |
+
# Convert to Eastern Time Zone
|
| 172 |
+
eastern_time = utc_time + timedelta(hours=-5)
|
| 173 |
+
# Format as string (YY-MM-DD HH:MM)
|
| 174 |
+
published_date = eastern_time.strftime('%m-%d-%y %H:%M')
|
| 175 |
+
|
| 176 |
+
import pandas as pd
|
| 177 |
+
# string dataframe?
|
| 178 |
+
df = pd.DataFrame([chat_transcript_copy])
|
| 179 |
+
notion_df.upload(df, 'https://www.notion.so/page-827360c361f347f7bfefcc6dfbd10e51', title=str(chat_transcript_copy), api_key=API_KEY)
|
| 180 |
|
| 181 |
return chat_transcript
|
| 182 |
|
|
|
|
| 204 |
|
| 205 |
|
| 206 |
|
|
|
|
|
|
|
| 207 |
# from transformers import pipeline, T5Tokenizer
|
| 208 |
# import pyttsx3
|
| 209 |
# import threading
|