Update functions.py
Browse files- functions.py +3 -4
functions.py
CHANGED
|
@@ -2,6 +2,7 @@ import whisper
|
|
| 2 |
import os
|
| 3 |
import random
|
| 4 |
import openai
|
|
|
|
| 5 |
import yt_dlp
|
| 6 |
from pytube import YouTube, extract
|
| 7 |
import pandas as pd
|
|
@@ -21,8 +22,6 @@ from bs4 import BeautifulSoup
|
|
| 21 |
import base64, time
|
| 22 |
from annotated_text import annotated_text
|
| 23 |
import pickle, math
|
| 24 |
-
import wikipedia
|
| 25 |
-
from pyvis.network import Network
|
| 26 |
import torch
|
| 27 |
from pydub import AudioSegment
|
| 28 |
from langchain.docstore.document import Document
|
|
@@ -57,7 +56,7 @@ from langchain.prompts import PromptTemplate
|
|
| 57 |
from langsmith import Client
|
| 58 |
|
| 59 |
client = Client()
|
| 60 |
-
|
| 61 |
nltk.download('punkt')
|
| 62 |
|
| 63 |
|
|
@@ -153,7 +152,7 @@ def load_whisper_api(audio):
|
|
| 153 |
|
| 154 |
'''Transcribe YT audio to text using Open AI API'''
|
| 155 |
file = open(audio, "rb")
|
| 156 |
-
transcript =
|
| 157 |
|
| 158 |
return transcript
|
| 159 |
|
|
|
|
| 2 |
import os
|
| 3 |
import random
|
| 4 |
import openai
|
| 5 |
+
from openai import OpenAI
|
| 6 |
import yt_dlp
|
| 7 |
from pytube import YouTube, extract
|
| 8 |
import pandas as pd
|
|
|
|
| 22 |
import base64, time
|
| 23 |
from annotated_text import annotated_text
|
| 24 |
import pickle, math
|
|
|
|
|
|
|
| 25 |
import torch
|
| 26 |
from pydub import AudioSegment
|
| 27 |
from langchain.docstore.document import Document
|
|
|
|
| 56 |
from langsmith import Client
|
| 57 |
|
| 58 |
client = Client()
|
| 59 |
+
openai_audio = OpenAI()
|
| 60 |
nltk.download('punkt')
|
| 61 |
|
| 62 |
|
|
|
|
| 152 |
|
| 153 |
'''Transcribe YT audio to text using Open AI API'''
|
| 154 |
file = open(audio, "rb")
|
| 155 |
+
transcript = openai_audio.audio.transcriptions.create("whisper-2", file)
|
| 156 |
|
| 157 |
return transcript
|
| 158 |
|