From 715d59784007002356333ce42a5bb25419586305 Mon Sep 17 00:00:00 2001
From: Serge Retkowsky <serge.retkowsky@microsoft.com>
Date: Mon, 11 Sep 2023 12:44:32 +0200
Subject: [PATCH] Add files via upload

---
 Utilities/Tokenizer.ipynb              | 1313 ++++++++++++++++++++++++
 Utilities/meeting_notes_from_audio.txt |   35 +
 2 files changed, 1348 insertions(+)
 create mode 100644 Utilities/Tokenizer.ipynb
 create mode 100644 Utilities/meeting_notes_from_audio.txt

diff --git a/Utilities/Tokenizer.ipynb b/Utilities/Tokenizer.ipynb
new file mode 100644
index 0000000..83ec1e2
--- /dev/null
+++ b/Utilities/Tokenizer.ipynb
@@ -0,0 +1,1313 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "bbeeb8bc",
+   "metadata": {},
+   "source": [
+    "# Tokenizer"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "eeefd124",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os\n",
+    "import openai\n",
+    "import tiktoken\n",
+    "\n",
+    "from dotenv import load_dotenv"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "023347ac",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "openai version = 0.28.0\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(\"openai version =\", openai.__version__)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "1eb66c65",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "load_dotenv(\"azure.env\")\n",
+    "\n",
+    "openai.api_key = os.getenv(\"OPENAI_API_KEY\")\n",
+    "openai.api_base = os.getenv(\"OPENAI_API_BASE\")\n",
+    "openai.api_type = \"azure\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "691e1935",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "encoding = tiktoken.get_encoding(\"cl100k_base\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "8905048a",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "<Encoding 'cl100k_base'>\n"
+     ]
+    }
+   ],
+   "source": [
+    "encoding = tiktoken.encoding_for_model(\"gpt-3.5-turbo\")\n",
+    "print(encoding)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "06547221",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "file_path = \"meeting_notes_from_audio.txt\""
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "09ea8601",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "try:\n",
+    "    with open(file_path, \"r\") as file:\n",
+    "        meeting_notes = file.read()\n",
+    "\n",
+    "except FileNotFoundError:\n",
+    "    print(f\"The file '{file_path}' was not found.\")\n",
+    "except Exception as e:\n",
+    "    print(f\"An error occurred: {str(e)}\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "74b82bf3",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[5159,\n",
+       " 6574,\n",
+       " 8554,\n",
+       " 15590,\n",
+       " 555,\n",
+       " 35219,\n",
+       " 5377,\n",
+       " 15592,\n",
+       " 323,\n",
+       " 35219,\n",
+       " 39841,\n",
+       " 3600,\n",
+       " 13,\n",
+       " 2696,\n",
+       " 25,\n",
+       " 220,\n",
+       " 806,\n",
+       " 96253,\n",
+       " 12,\n",
+       " 2366,\n",
+       " 18,\n",
+       " 220,\n",
+       " 806,\n",
+       " 25,\n",
+       " 2970,\n",
+       " 25,\n",
+       " 2491,\n",
+       " 271,\n",
+       " 19791,\n",
+       " 1473,\n",
+       " 37,\n",
+       " 396,\n",
+       " 4842,\n",
+       " 12623,\n",
+       " 30037,\n",
+       " 1047,\n",
+       " 264,\n",
+       " 6992,\n",
+       " 2132,\n",
+       " 8502,\n",
+       " 315,\n",
+       " 220,\n",
+       " 2366,\n",
+       " 18,\n",
+       " 449,\n",
+       " 264,\n",
+       " 13254,\n",
+       " 315,\n",
+       " 220,\n",
+       " 6549,\n",
+       " 3610,\n",
+       " 323,\n",
+       " 264,\n",
+       " 20547,\n",
+       " 11626,\n",
+       " 4850,\n",
+       " 315,\n",
+       " 220,\n",
+       " 2970,\n",
+       " 14697,\n",
+       " 11205,\n",
+       " 4272,\n",
+       " 8070,\n",
+       " 574,\n",
+       " 220,\n",
+       " 845,\n",
+       " 3610,\n",
+       " 11,\n",
+       " 264,\n",
+       " 5199,\n",
+       " 5376,\n",
+       " 505,\n",
+       " 279,\n",
+       " 220,\n",
+       " 605,\n",
+       " 3610,\n",
+       " 3970,\n",
+       " 304,\n",
+       " 279,\n",
+       " 1890,\n",
+       " 8502,\n",
+       " 315,\n",
+       " 279,\n",
+       " 3766,\n",
+       " 1060,\n",
+       " 13,\n",
+       " 23212,\n",
+       " 11,\n",
+       " 435,\n",
+       " 396,\n",
+       " 4842,\n",
+       " 12623,\n",
+       " 706,\n",
+       " 85957,\n",
+       " 872,\n",
+       " 9513,\n",
+       " 46128,\n",
+       " 34919,\n",
+       " 20136,\n",
+       " 323,\n",
+       " 29091,\n",
+       " 304,\n",
+       " 48197,\n",
+       " 22359,\n",
+       " 13166,\n",
+       " 27460,\n",
+       " 311,\n",
+       " 18885,\n",
+       " 5326,\n",
+       " 60684,\n",
+       " 291,\n",
+       " 4780,\n",
+       " 13,\n",
+       " 24296,\n",
+       " 11,\n",
+       " 814,\n",
+       " 617,\n",
+       " 3970,\n",
+       " 17808,\n",
+       " 1217,\n",
+       " 6650,\n",
+       " 449,\n",
+       " 264,\n",
+       " 445,\n",
+       " 16027,\n",
+       " 11547,\n",
+       " 1741,\n",
+       " 11595,\n",
+       " 315,\n",
+       " 220,\n",
+       " 18,\n",
+       " 13,\n",
+       " 20,\n",
+       " 13689,\n",
+       " 323,\n",
+       " 872,\n",
+       " 18057,\n",
+       " 369,\n",
+       " 279,\n",
+       " 5108,\n",
+       " 8502,\n",
+       " 374,\n",
+       " 6928,\n",
+       " 449,\n",
+       " 220,\n",
+       " 23,\n",
+       " 4,\n",
+       " 8502,\n",
+       " 29352,\n",
+       " 58414,\n",
+       " 6650,\n",
+       " 13,\n",
+       " 17830,\n",
+       " 11,\n",
+       " 872,\n",
+       " 435,\n",
+       " 396,\n",
+       " 4842,\n",
+       " 41164,\n",
+       " 374,\n",
+       " 743,\n",
+       " 311,\n",
+       " 6929,\n",
+       " 459,\n",
+       " 67992,\n",
+       " 902,\n",
+       " 374,\n",
+       " 3685,\n",
+       " 311,\n",
+       " 4933,\n",
+       " 220,\n",
+       " 1049,\n",
+       " 3610,\n",
+       " 382,\n",
+       " 6334,\n",
+       " 21387,\n",
+       " 512,\n",
+       " 12,\n",
+       " 435,\n",
+       " 396,\n",
+       " 4842,\n",
+       " 12623,\n",
+       " 30037,\n",
+       " 1047,\n",
+       " 264,\n",
+       " 3831,\n",
+       " 2132,\n",
+       " 8502,\n",
+       " 449,\n",
+       " 264,\n",
+       " 220,\n",
+       " 914,\n",
+       " 4,\n",
+       " 44188,\n",
+       " 56,\n",
+       " 13254,\n",
+       " 5376,\n",
+       " 198,\n",
+       " 12,\n",
+       " 43907,\n",
+       " 11626,\n",
+       " 4850,\n",
+       " 574,\n",
+       " 220,\n",
+       " 2970,\n",
+       " 4,\n",
+       " 4245,\n",
+       " 311,\n",
+       " 2853,\n",
+       " 92126,\n",
+       " 198,\n",
+       " 12,\n",
+       " 13315,\n",
+       " 300,\n",
+       " 85007,\n",
+       " 311,\n",
+       " 220,\n",
+       " 1806,\n",
+       " 13,\n",
+       " 20,\n",
+       " 3610,\n",
+       " 67371,\n",
+       " 311,\n",
+       " 264,\n",
+       " 220,\n",
+       " 966,\n",
+       " 4,\n",
+       " 469,\n",
+       " 21587,\n",
+       " 6486,\n",
+       " 4850,\n",
+       " 198,\n",
+       " 12,\n",
+       " 9558,\n",
+       " 8070,\n",
+       " 315,\n",
+       " 220,\n",
+       " 845,\n",
+       " 3610,\n",
+       " 11,\n",
+       " 709,\n",
+       " 505,\n",
+       " 220,\n",
+       " 605,\n",
+       " 3610,\n",
+       " 198,\n",
+       " 12,\n",
+       " 10884,\n",
+       " 2686,\n",
+       " 481,\n",
+       " 3157,\n",
+       " 14264,\n",
+       " 4245,\n",
+       " 311,\n",
+       " 17626,\n",
+       " 2027,\n",
+       " 1584,\n",
+       " 198,\n",
+       " 12,\n",
+       " 13038,\n",
+       " 291,\n",
+       " 304,\n",
+       " 9513,\n",
+       " 46128,\n",
+       " 34919,\n",
+       " 323,\n",
+       " 13166,\n",
+       " 27460,\n",
+       " 198,\n",
+       " 12,\n",
+       " 31946,\n",
+       " 11071,\n",
+       " 12032,\n",
+       " 315,\n",
+       " 220,\n",
+       " 16,\n",
+       " 13,\n",
+       " 20,\n",
+       " 7239,\n",
+       " 449,\n",
+       " 25452,\n",
+       " 315,\n",
+       " 220,\n",
+       " 5067,\n",
+       " 3610,\n",
+       " 198,\n",
+       " 12,\n",
+       " 58654,\n",
+       " 311,\n",
+       " 25452,\n",
+       " 11595,\n",
+       " 315,\n",
+       " 220,\n",
+       " 16,\n",
+       " 13,\n",
+       " 20,\n",
+       " 720,\n",
+       " 12,\n",
+       " 62697,\n",
+       " 17808,\n",
+       " 1217,\n",
+       " 6650,\n",
+       " 323,\n",
+       " 4827,\n",
+       " 6130,\n",
+       " 24279,\n",
+       " 2853,\n",
+       " 198,\n",
+       " 12,\n",
+       " 445,\n",
+       " 16027,\n",
+       " 356,\n",
+       " 1741,\n",
+       " 11595,\n",
+       " 315,\n",
+       " 220,\n",
+       " 18,\n",
+       " 13,\n",
+       " 20,\n",
+       " 14062,\n",
+       " 12,\n",
+       " 5273,\n",
+       " 520,\n",
+       " 5326,\n",
+       " 1646,\n",
+       " 304,\n",
+       " 2035,\n",
+       " 198,\n",
+       " 12,\n",
+       " 56775,\n",
+       " 369,\n",
+       " 5108,\n",
+       " 8502,\n",
+       " 315,\n",
+       " 220,\n",
+       " 8878,\n",
+       " 3610,\n",
+       " 13254,\n",
+       " 198,\n",
+       " 12,\n",
+       " 3216,\n",
+       " 5065,\n",
+       " 67992,\n",
+       " 315,\n",
+       " 11728,\n",
+       " 12623,\n",
+       " 198,\n",
+       " 12,\n",
+       " 56775,\n",
+       " 19738,\n",
+       " 6650,\n",
+       " 15174,\n",
+       " 271,\n",
+       " 2573,\n",
+       " 19974,\n",
+       " 1473,\n",
+       " 16,\n",
+       " 13,\n",
+       " 13038,\n",
+       " 220,\n",
+       " 914,\n",
+       " 11,\n",
+       " 931,\n",
+       " 11,\n",
+       " 931,\n",
+       " 304,\n",
+       " 48197,\n",
+       " 22359,\n",
+       " 13166,\n",
+       " 27460,\n",
+       " 13,\n",
+       " 720,\n",
+       " 17,\n",
+       " 13,\n",
+       " 423,\n",
+       " 1986,\n",
+       " 1463,\n",
+       " 9513,\n",
+       " 22126,\n",
+       " 34919,\n",
+       " 20136,\n",
+       " 13,\n",
+       " 720,\n",
+       " 18,\n",
+       " 13,\n",
+       " 13038,\n",
+       " 17345,\n",
+       " 304,\n",
+       " 52508,\n",
+       " 1534,\n",
+       " 11897,\n",
+       " 30255,\n",
+       " 323,\n",
+       " 20658,\n",
+       " 21515,\n",
+       " 46128,\n",
+       " 34919,\n",
+       " 13,\n",
+       " 720,\n",
+       " 19,\n",
+       " 13,\n",
+       " 51241,\n",
+       " 1579,\n",
+       " 7692,\n",
+       " 19523,\n",
+       " 2027,\n",
+       " 1584,\n",
+       " 13,\n",
+       " 720,\n",
+       " 20,\n",
+       " 13,\n",
+       " 8000,\n",
+       " 502,\n",
+       " 4997,\n",
+       " 78,\n",
+       " 37713,\n",
+       " 5452,\n",
+       " 13,\n",
+       " 720,\n",
+       " 21,\n",
+       " 13,\n",
+       " 32175,\n",
+       " 907,\n",
+       " 520,\n",
+       " 5326,\n",
+       " 1646,\n",
+       " 449,\n",
+       " 264,\n",
+       " 220,\n",
+       " 1484,\n",
+       " 4,\n",
+       " 12410,\n",
+       " 2237,\n",
+       " 13,\n",
+       " 720,\n",
+       " 22,\n",
+       " 13,\n",
+       " 86380,\n",
+       " 15692,\n",
+       " 5603,\n",
+       " 311,\n",
+       " 18646,\n",
+       " 33164,\n",
+       " 13,\n",
+       " 720,\n",
+       " 23,\n",
+       " 13,\n",
+       " 24083,\n",
+       " 435,\n",
+       " 396,\n",
+       " 4842,\n",
+       " 41164,\n",
+       " 11728,\n",
+       " 12623,\n",
+       " 67992,\n",
+       " 311,\n",
+       " 4933,\n",
+       " 220,\n",
+       " 1049,\n",
+       " 3610,\n",
+       " 13,\n",
+       " 720,\n",
+       " 24,\n",
+       " 13,\n",
+       " 32175,\n",
+       " 19738,\n",
+       " 6650,\n",
+       " 15174,\n",
+       " 382,\n",
+       " 791,\n",
+       " 27065,\n",
+       " 315,\n",
+       " 279,\n",
+       " 1495,\n",
+       " 374,\n",
+       " 8965,\n",
+       " 6928,\n",
+       " 13,\n",
+       " 578,\n",
+       " 4221,\n",
+       " 1511,\n",
+       " 390,\n",
+       " 50369,\n",
+       " 36232,\n",
+       " 323,\n",
+       " 54508,\n",
+       " 11,\n",
+       " 39686,\n",
+       " 279,\n",
+       " 16358,\n",
+       " 6650,\n",
+       " 323,\n",
+       " 6020,\n",
+       " 5178,\n",
+       " 315,\n",
+       " 279,\n",
+       " 2883,\n",
+       " 13,\n",
+       " 578,\n",
+       " 12432,\n",
+       " 1101,\n",
+       " 61120,\n",
+       " 46135,\n",
+       " 311,\n",
+       " 41777,\n",
+       " 11,\n",
+       " 902,\n",
+       " 15151,\n",
+       " 264,\n",
+       " 6928,\n",
+       " 19451,\n",
+       " 13,\n",
+       " 23212,\n",
+       " 11,\n",
+       " 279,\n",
+       " 2317,\n",
+       " 315,\n",
+       " 279,\n",
+       " 10430,\n",
+       " 374,\n",
+       " 6928,\n",
+       " 11,\n",
+       " 449,\n",
+       " 279,\n",
+       " 12432,\n",
+       " 38787,\n",
+       " 264,\n",
+       " 6992,\n",
+       " 8502,\n",
+       " 323,\n",
+       " 264,\n",
+       " 6992,\n",
+       " 14827,\n",
+       " 67992,\n",
+       " 13]"
+      ]
+     },
+     "execution_count": 8,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "tokens_integer = encoding.encode(meeting_notes)\n",
+    "tokens_integer"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 9,
+   "id": "fbe71cce",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "535 is the number of tokens in my text\n"
+     ]
+    }
+   ],
+   "source": [
+    "print(f\"{len(tokens_integer)} is the number of tokens in my text\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "7e44a2fe",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "[b'My',\n",
+       " b' meeting',\n",
+       " b' notes',\n",
+       " b' processed',\n",
+       " b' by',\n",
+       " b' Azure',\n",
+       " b' Open',\n",
+       " b' AI',\n",
+       " b' and',\n",
+       " b' Azure',\n",
+       " b' Speech',\n",
+       " b' services',\n",
+       " b'.',\n",
+       " b' Date',\n",
+       " b':',\n",
+       " b' ',\n",
+       " b'11',\n",
+       " b'-Sep',\n",
+       " b'-',\n",
+       " b'202',\n",
+       " b'3',\n",
+       " b' ',\n",
+       " b'11',\n",
+       " b':',\n",
+       " b'58',\n",
+       " b':',\n",
+       " b'49',\n",
+       " b'\\n\\n',\n",
+       " b'Summary',\n",
+       " b':\\n\\n',\n",
+       " b'F',\n",
+       " b'int',\n",
+       " b'ech',\n",
+       " b' Plus',\n",
+       " b' Sync',\n",
+       " b' had',\n",
+       " b' a',\n",
+       " b' successful',\n",
+       " b' second',\n",
+       " b' quarter',\n",
+       " b' of',\n",
+       " b' ',\n",
+       " b'202',\n",
+       " b'3',\n",
+       " b' with',\n",
+       " b' a',\n",
+       " b' revenue',\n",
+       " b' of',\n",
+       " b' ',\n",
+       " b'125',\n",
+       " b' million',\n",
+       " b' and',\n",
+       " b' a',\n",
+       " b' gross',\n",
+       " b' profit',\n",
+       " b' margin',\n",
+       " b' of',\n",
+       " b' ',\n",
+       " b'58',\n",
+       " b'%.',\n",
+       " b' Their',\n",
+       " b' net',\n",
+       " b' income',\n",
+       " b' was',\n",
+       " b' ',\n",
+       " b'16',\n",
+       " b' million',\n",
+       " b',',\n",
+       " b' a',\n",
+       " b' significant',\n",
+       " b' increase',\n",
+       " b' from',\n",
+       " b' the',\n",
+       " b' ',\n",
+       " b'10',\n",
+       " b' million',\n",
+       " b' seen',\n",
+       " b' in',\n",
+       " b' the',\n",
+       " b' same',\n",
+       " b' quarter',\n",
+       " b' of',\n",
+       " b' the',\n",
+       " b' previous',\n",
+       " b' year',\n",
+       " b'.',\n",
+       " b' Additionally',\n",
+       " b',',\n",
+       " b' F',\n",
+       " b'int',\n",
+       " b'ech',\n",
+       " b' Plus',\n",
+       " b' has',\n",
+       " b' diversified',\n",
+       " b' their',\n",
+       " b' asset',\n",
+       " b'-backed',\n",
+       " b' securities',\n",
+       " b' portfolio',\n",
+       " b' and',\n",
+       " b' invested',\n",
+       " b' in',\n",
+       " b' AAA',\n",
+       " b' rated',\n",
+       " b' corporate',\n",
+       " b' bonds',\n",
+       " b' to',\n",
+       " b' enhance',\n",
+       " b' risk',\n",
+       " b'-adjust',\n",
+       " b'ed',\n",
+       " b' returns',\n",
+       " b'.',\n",
+       " b' Furthermore',\n",
+       " b',',\n",
+       " b' they',\n",
+       " b' have',\n",
+       " b' seen',\n",
+       " b' organic',\n",
+       " b' user',\n",
+       " b' growth',\n",
+       " b' with',\n",
+       " b' a',\n",
+       " b' L',\n",
+       " b'TV',\n",
+       " b'/C',\n",
+       " b'AC',\n",
+       " b' ratio',\n",
+       " b' of',\n",
+       " b' ',\n",
+       " b'3',\n",
+       " b'.',\n",
+       " b'5',\n",
+       " b'%,',\n",
+       " b' and',\n",
+       " b' their',\n",
+       " b' forecast',\n",
+       " b' for',\n",
+       " b' the',\n",
+       " b' coming',\n",
+       " b' quarter',\n",
+       " b' is',\n",
+       " b' positive',\n",
+       " b' with',\n",
+       " b' ',\n",
+       " b'8',\n",
+       " b'%',\n",
+       " b' quarter',\n",
+       " b'-over',\n",
+       " b'-quarter',\n",
+       " b' growth',\n",
+       " b'.',\n",
+       " b' Finally',\n",
+       " b',',\n",
+       " b' their',\n",
+       " b' F',\n",
+       " b'int',\n",
+       " b'ech',\n",
+       " b' subsidiary',\n",
+       " b' is',\n",
+       " b' set',\n",
+       " b' to',\n",
+       " b' conduct',\n",
+       " b' an',\n",
+       " b' IPO',\n",
+       " b' which',\n",
+       " b' is',\n",
+       " b' expected',\n",
+       " b' to',\n",
+       " b' raise',\n",
+       " b' ',\n",
+       " b'200',\n",
+       " b' million',\n",
+       " b'.\\n\\n',\n",
+       " b'Main',\n",
+       " b' Points',\n",
+       " b':\\n',\n",
+       " b'-',\n",
+       " b' F',\n",
+       " b'int',\n",
+       " b'ech',\n",
+       " b' Plus',\n",
+       " b' Sync',\n",
+       " b' had',\n",
+       " b' a',\n",
+       " b' strong',\n",
+       " b' second',\n",
+       " b' quarter',\n",
+       " b' with',\n",
+       " b' a',\n",
+       " b' ',\n",
+       " b'25',\n",
+       " b'%',\n",
+       " b' Yo',\n",
+       " b'Y',\n",
+       " b' revenue',\n",
+       " b' increase',\n",
+       " b'\\n',\n",
+       " b'-',\n",
+       " b' Gross',\n",
+       " b' profit',\n",
+       " b' margin',\n",
+       " b' was',\n",
+       " b' ',\n",
+       " b'58',\n",
+       " b'%',\n",
+       " b' due',\n",
+       " b' to',\n",
+       " b' cost',\n",
+       " b' efficiencies',\n",
+       " b'\\n',\n",
+       " b'-',\n",
+       " b' Bet',\n",
+       " b'as',\n",
+       " b' surged',\n",
+       " b' to',\n",
+       " b' ',\n",
+       " b'37',\n",
+       " b'.',\n",
+       " b'5',\n",
+       " b' million',\n",
+       " b' translating',\n",
+       " b' to',\n",
+       " b' a',\n",
+       " b' ',\n",
+       " b'30',\n",
+       " b'%',\n",
+       " b' E',\n",
+       " b'BIT',\n",
+       " b'DA',\n",
+       " b' margin',\n",
+       " b'\\n',\n",
+       " b'-',\n",
+       " b' Net',\n",
+       " b' income',\n",
+       " b' of',\n",
+       " b' ',\n",
+       " b'16',\n",
+       " b' million',\n",
+       " b',',\n",
+       " b' up',\n",
+       " b' from',\n",
+       " b' ',\n",
+       " b'10',\n",
+       " b' million',\n",
+       " b'\\n',\n",
+       " b'-',\n",
+       " b' Total',\n",
+       " b' address',\n",
+       " b'able',\n",
+       " b' market',\n",
+       " b' grew',\n",
+       " b' due',\n",
+       " b' to',\n",
+       " b' expanded',\n",
+       " b' product',\n",
+       " b' line',\n",
+       " b'\\n',\n",
+       " b'-',\n",
+       " b' Invest',\n",
+       " b'ed',\n",
+       " b' in',\n",
+       " b' asset',\n",
+       " b'-backed',\n",
+       " b' securities',\n",
+       " b' and',\n",
+       " b' corporate',\n",
+       " b' bonds',\n",
+       " b'\\n',\n",
+       " b'-',\n",
+       " b' Balance',\n",
+       " b' sheet',\n",
+       " b' assets',\n",
+       " b' of',\n",
+       " b' ',\n",
+       " b'1',\n",
+       " b'.',\n",
+       " b'5',\n",
+       " b' billion',\n",
+       " b' with',\n",
+       " b' equity',\n",
+       " b' of',\n",
+       " b' ',\n",
+       " b'600',\n",
+       " b' million',\n",
+       " b'\\n',\n",
+       " b'-',\n",
+       " b' Debt',\n",
+       " b' to',\n",
+       " b' equity',\n",
+       " b' ratio',\n",
+       " b' of',\n",
+       " b' ',\n",
+       " b'1',\n",
+       " b'.',\n",
+       " b'5',\n",
+       " b' \\n',\n",
+       " b'-',\n",
+       " b' Increased',\n",
+       " b' organic',\n",
+       " b' user',\n",
+       " b' growth',\n",
+       " b' and',\n",
+       " b' lower',\n",
+       " b' customer',\n",
+       " b' acquisition',\n",
+       " b' cost',\n",
+       " b'\\n',\n",
+       " b'-',\n",
+       " b' L',\n",
+       " b'TV',\n",
+       " b' C',\n",
+       " b'AC',\n",
+       " b' ratio',\n",
+       " b' of',\n",
+       " b' ',\n",
+       " b'3',\n",
+       " b'.',\n",
+       " b'5',\n",
+       " b'%\\n',\n",
+       " b'-',\n",
+       " b' Value',\n",
+       " b' at',\n",
+       " b' risk',\n",
+       " b' model',\n",
+       " b' in',\n",
+       " b' place',\n",
+       " b'\\n',\n",
+       " b'-',\n",
+       " b' Forecast',\n",
+       " b' for',\n",
+       " b' coming',\n",
+       " b' quarter',\n",
+       " b' of',\n",
+       " b' ',\n",
+       " b'135',\n",
+       " b' million',\n",
+       " b' revenue',\n",
+       " b'\\n',\n",
+       " b'-',\n",
+       " b' Up',\n",
+       " b'coming',\n",
+       " b' IPO',\n",
+       " b' of',\n",
+       " b' Pay',\n",
+       " b' Plus',\n",
+       " b'\\n',\n",
+       " b'-',\n",
+       " b' Forecast',\n",
+       " b' aggressive',\n",
+       " b' growth',\n",
+       " b' strategies',\n",
+       " b'\\n\\n',\n",
+       " b'Action',\n",
+       " b' Items',\n",
+       " b':\\n\\n',\n",
+       " b'1',\n",
+       " b'.',\n",
+       " b' Invest',\n",
+       " b' ',\n",
+       " b'25',\n",
+       " b',',\n",
+       " b'000',\n",
+       " b',',\n",
+       " b'000',\n",
+       " b' in',\n",
+       " b' AAA',\n",
+       " b' rated',\n",
+       " b' corporate',\n",
+       " b' bonds',\n",
+       " b'.',\n",
+       " b' \\n',\n",
+       " b'2',\n",
+       " b'.',\n",
+       " b' D',\n",
+       " b'ivers',\n",
+       " b'ify',\n",
+       " b' asset',\n",
+       " b' backed',\n",
+       " b' securities',\n",
+       " b' portfolio',\n",
+       " b'.',\n",
+       " b' \\n',\n",
+       " b'3',\n",
+       " b'.',\n",
+       " b' Invest',\n",
+       " b' heavily',\n",
+       " b' in',\n",
+       " b' collateral',\n",
+       " b'ized',\n",
+       " b' debt',\n",
+       " b' obligations',\n",
+       " b' and',\n",
+       " b' residential',\n",
+       " b' mortgage',\n",
+       " b'-backed',\n",
+       " b' securities',\n",
+       " b'.',\n",
+       " b' \\n',\n",
+       " b'4',\n",
+       " b'.',\n",
+       " b' Expand',\n",
+       " b' high',\n",
+       " b' yield',\n",
+       " b' savings',\n",
+       " b' product',\n",
+       " b' line',\n",
+       " b'.',\n",
+       " b' \\n',\n",
+       " b'5',\n",
+       " b'.',\n",
+       " b' Develop',\n",
+       " b' new',\n",
+       " b' Rob',\n",
+       " b'o',\n",
+       " b' advisor',\n",
+       " b' platform',\n",
+       " b'.',\n",
+       " b' \\n',\n",
+       " b'6',\n",
+       " b'.',\n",
+       " b' Implement',\n",
+       " b' value',\n",
+       " b' at',\n",
+       " b' risk',\n",
+       " b' model',\n",
+       " b' with',\n",
+       " b' a',\n",
+       " b' ',\n",
+       " b'99',\n",
+       " b'%',\n",
+       " b' confidence',\n",
+       " b' level',\n",
+       " b'.',\n",
+       " b' \\n',\n",
+       " b'7',\n",
+       " b'.',\n",
+       " b' Adopt',\n",
+       " b' conservative',\n",
+       " b' approach',\n",
+       " b' to',\n",
+       " b' managing',\n",
+       " b' leverage',\n",
+       " b'.',\n",
+       " b' \\n',\n",
+       " b'8',\n",
+       " b'.',\n",
+       " b' Launch',\n",
+       " b' F',\n",
+       " b'int',\n",
+       " b'ech',\n",
+       " b' subsidiary',\n",
+       " b' Pay',\n",
+       " b' Plus',\n",
+       " b' IPO',\n",
+       " b' to',\n",
+       " b' raise',\n",
+       " b' ',\n",
+       " b'200',\n",
+       " b' million',\n",
+       " b'.',\n",
+       " b' \\n',\n",
+       " b'9',\n",
+       " b'.',\n",
+       " b' Implement',\n",
+       " b' aggressive',\n",
+       " b' growth',\n",
+       " b' strategies',\n",
+       " b'.\\n\\n',\n",
+       " b'The',\n",
+       " b' sentiment',\n",
+       " b' of',\n",
+       " b' the',\n",
+       " b' text',\n",
+       " b' is',\n",
+       " b' generally',\n",
+       " b' positive',\n",
+       " b'.',\n",
+       " b' The',\n",
+       " b' language',\n",
+       " b' used',\n",
+       " b' con',\n",
+       " b'veys',\n",
+       " b' enthusiasm',\n",
+       " b' and',\n",
+       " b' optimism',\n",
+       " b',',\n",
+       " b' highlighting',\n",
+       " b' the',\n",
+       " b' impressive',\n",
+       " b' growth',\n",
+       " b' and',\n",
+       " b' financial',\n",
+       " b' performance',\n",
+       " b' of',\n",
+       " b' the',\n",
+       " b' company',\n",
+       " b'.',\n",
+       " b' The',\n",
+       " b' CEO',\n",
+       " b' also',\n",
+       " b' expresses',\n",
+       " b' gratitude',\n",
+       " b' to',\n",
+       " b' shareholders',\n",
+       " b',',\n",
+       " b' which',\n",
+       " b' indicates',\n",
+       " b' a',\n",
+       " b' positive',\n",
+       " b' attitude',\n",
+       " b'.',\n",
+       " b' Additionally',\n",
+       " b',',\n",
+       " b' the',\n",
+       " b' context',\n",
+       " b' of',\n",
+       " b' the',\n",
+       " b' discussion',\n",
+       " b' is',\n",
+       " b' positive',\n",
+       " b',',\n",
+       " b' with',\n",
+       " b' the',\n",
+       " b' CEO',\n",
+       " b' announcing',\n",
+       " b' a',\n",
+       " b' successful',\n",
+       " b' quarter',\n",
+       " b' and',\n",
+       " b' a',\n",
+       " b' successful',\n",
+       " b' upcoming',\n",
+       " b' IPO',\n",
+       " b'.']"
+      ]
+     },
+     "execution_count": 10,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "tokens_string = [encoding.decode_single_token_bytes(token) for token in tokens_integer]\n",
+    "tokens_string"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "3c8cdf73",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "1 is the number of token included in role\n",
+      "14 is the number of token included in content\n",
+      "22 number of tokens to be sent in our request\n"
+     ]
+    }
+   ],
+   "source": [
+    "message = [\n",
+    "    {\n",
+    "        \"role\": \"user\",\n",
+    "        \"content\": \"Explain to me how tokenization is working in OpenAI models?\",\n",
+    "    }\n",
+    "]\n",
+    "\n",
+    "tokens_per_message = 4\n",
+    "\n",
+    "num_tokens = 0\n",
+    "num_tokens += tokens_per_message\n",
+    "\n",
+    "for key, value in message[0].items():\n",
+    "    text = value\n",
+    "    num_tokens += len(encoding.encode(value))\n",
+    "    print(f\"{len(encoding.encode(value))} is the number of token included in {key}\")\n",
+    "\n",
+    "num_tokens += 3\n",
+    "# every reply is primed with <|start|>assistant<|message|>\n",
+    "\n",
+    "print(f\"{num_tokens} number of tokens to be sent in our request\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "c9236a00",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3.10 - SDK v2",
+   "language": "python",
+   "name": "python310-sdkv2"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.10"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}
diff --git a/Utilities/meeting_notes_from_audio.txt b/Utilities/meeting_notes_from_audio.txt
new file mode 100644
index 0000000..5cdfec7
--- /dev/null
+++ b/Utilities/meeting_notes_from_audio.txt
@@ -0,0 +1,35 @@
+My meeting notes processed by Azure Open AI and Azure Speech services. Date: 11-Sep-2023 11:58:49
+
+Summary:
+
+Fintech Plus Sync had a successful second quarter of 2023 with a revenue of 125 million and a gross profit margin of 58%. Their net income was 16 million, a significant increase from the 10 million seen in the same quarter of the previous year. Additionally, Fintech Plus has diversified their asset-backed securities portfolio and invested in AAA rated corporate bonds to enhance risk-adjusted returns. Furthermore, they have seen organic user growth with a LTV/CAC ratio of 3.5%, and their forecast for the coming quarter is positive with 8% quarter-over-quarter growth. Finally, their Fintech subsidiary is set to conduct an IPO which is expected to raise 200 million.
+
+Main Points:
+- Fintech Plus Sync had a strong second quarter with a 25% YoY revenue increase
+- Gross profit margin was 58% due to cost efficiencies
+- Betas surged to 37.5 million translating to a 30% EBITDA margin
+- Net income of 16 million, up from 10 million
+- Total addressable market grew due to expanded product line
+- Invested in asset-backed securities and corporate bonds
+- Balance sheet assets of 1.5 billion with equity of 600 million
+- Debt to equity ratio of 1.5 
+- Increased organic user growth and lower customer acquisition cost
+- LTV CAC ratio of 3.5%
+- Value at risk model in place
+- Forecast for coming quarter of 135 million revenue
+- Upcoming IPO of Pay Plus
+- Forecast aggressive growth strategies
+
+Action Items:
+
+1. Invest 25,000,000 in AAA rated corporate bonds. 
+2. Diversify asset backed securities portfolio. 
+3. Invest heavily in collateralized debt obligations and residential mortgage-backed securities. 
+4. Expand high yield savings product line. 
+5. Develop new Robo advisor platform. 
+6. Implement value at risk model with a 99% confidence level. 
+7. Adopt conservative approach to managing leverage. 
+8. Launch Fintech subsidiary Pay Plus IPO to raise 200 million. 
+9. Implement aggressive growth strategies.
+
+The sentiment of the text is generally positive. The language used conveys enthusiasm and optimism, highlighting the impressive growth and financial performance of the company. The CEO also expresses gratitude to shareholders, which indicates a positive attitude. Additionally, the context of the discussion is positive, with the CEO announcing a successful quarter and a successful upcoming IPO.
\ No newline at end of file