Spaces:
Sleeping
Sleeping
YouTubeTranscriptApi
Browse files- app.py +5 -1
- requirements.txt +2 -1
app.py
CHANGED
|
@@ -6,6 +6,8 @@ from docx import Document
|
|
| 6 |
import os
|
| 7 |
from openai import OpenAI
|
| 8 |
import json
|
|
|
|
|
|
|
| 9 |
|
| 10 |
OPEN_AI_KEY = os.getenv("OPEN_AI_KEY")
|
| 11 |
client = OpenAI(api_key=OPEN_AI_KEY)
|
|
@@ -50,7 +52,9 @@ def docx_to_text(file):
|
|
| 50 |
def process_youtube_link(link):
|
| 51 |
# 使用 YouTube API 获取逐字稿
|
| 52 |
# 假设您已经获取了 YouTube 视频的逐字稿并存储在变量 `transcript` 中
|
| 53 |
-
|
|
|
|
|
|
|
| 54 |
|
| 55 |
# 基于逐字稿生成其他所需的输出
|
| 56 |
questions = generate_questions(transcript)
|
|
|
|
| 6 |
import os
|
| 7 |
from openai import OpenAI
|
| 8 |
import json
|
| 9 |
+
from youtube_transcript_api import YouTubeTranscriptApi
|
| 10 |
+
|
| 11 |
|
| 12 |
OPEN_AI_KEY = os.getenv("OPEN_AI_KEY")
|
| 13 |
client = OpenAI(api_key=OPEN_AI_KEY)
|
|
|
|
| 52 |
def process_youtube_link(link):
|
| 53 |
# 使用 YouTube API 获取逐字稿
|
| 54 |
# 假设您已经获取了 YouTube 视频的逐字稿并存储在变量 `transcript` 中
|
| 55 |
+
video_id = link.split("=")[-1]
|
| 56 |
+
transcript = YouTubeTranscriptApi.get_transcript(video_id, languages=['zh-TW'])
|
| 57 |
+
|
| 58 |
|
| 59 |
# 基于逐字稿生成其他所需的输出
|
| 60 |
questions = generate_questions(transcript)
|
requirements.txt
CHANGED
|
@@ -3,4 +3,5 @@ pandas
|
|
| 3 |
openai>=1.0.0
|
| 4 |
requests
|
| 5 |
beautifulsoup4
|
| 6 |
-
python-docx
|
|
|
|
|
|
| 3 |
openai>=1.0.0
|
| 4 |
requests
|
| 5 |
beautifulsoup4
|
| 6 |
+
python-docx
|
| 7 |
+
youtube-transcript-api
|