You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: readme.md
+28-12Lines changed: 28 additions & 12 deletions
Original file line number
Diff line number
Diff line change
@@ -1,5 +1,6 @@
1
1
# YouTube Transcript Generator
2
-
[](https://colab.research.google.com/github/therohitdas/Youtube-Transcript-Generator/blob/main/main.ipynb)
2
+
3
+
[](https://colab.research.google.com/github/therohitdas/Youtube-Transcript-Generator/blob/main/main.ipynb)
@@ -10,6 +11,7 @@ The YouTube Transcript Generator is a powerful tool designed to streamline the p
10
11
### How It Can Help 🚀
11
12
12
13
This tool is particularly useful for:
14
+
13
15
-**Note Taking:** Quickly convert YouTube videos into text format for easy note-taking.
14
16
-**Content Analysis:** Analyze and derive insights from video content by converting it into text data.
15
17
-**Chat Bot Training:** Use the generated transcripts to train chatbots, such as ChatGPT, for natural language understanding.
@@ -26,47 +28,57 @@ This tool is particularly useful for:
26
28
27
29
## Environment Variables 🌐
28
30
29
-
-`GOOGLE_API_KEY`: Set up your Google API key for video information retrieval. You will need to create a Project in the Google Cloud for this and enable the YouTube v3 API. This is optional, if you don't add it, the chapters will not be added.
31
+
-`YOUTUBE_API_KEY`: Set up your Google API key for video information retrieval. You will need to create a Project in the Google Cloud for this and enable the YouTube v3 API. This is optional, if you don't add it, the chapters will not be added.
30
32
31
33
## Script Parameters 📜
34
+
32
35
When running the script locally, you can pass these parameters to the script:
33
36
34
37
### Positional Argument:
38
+
35
39
-`url`: YouTube video URL
36
40
37
41
### Optional Arguments:
42
+
38
43
-`-h, --help`: Show the help message and exit
39
44
-`-l LANGUAGE, --language LANGUAGE`: Language for the transcript (default: en)
To run this project in Google Colab, follow these steps:
48
-
1. Open the [Google Colab notebook](https://colab.research.google.com/github/therohitdas/Youtube-Transcript-Generator/blob/main/main.ipynb).
49
-
2. Add Google's Project API key to the secrets tab under this key: `GOOGLE_API_KEY` and toggle notebook access to on.
50
-
3. Edit the variables in the second last cell.
51
-
4. Go to Runtime > Change Runtime Type and select any GPU type. If you use CPU, the output for punctuated transcript will take some minutes to complete (around 1 minute per 10-minute video)
52
-
5. Change the values in the third last cell to include your URL etc.
54
+
55
+
1. Open the [Google Colab Notebook](https://colab.research.google.com/github/therohitdas/Youtube-Transcript-Generator/blob/main/main.ipynb).
56
+
2. Add Google's Project API key to the secrets tab under this key: `YOUTUBE_API_KEY` and toggle notebook access to on.
57
+
3. Go to Runtime > Change Runtime Type and select T4 GPU type. If you use CPU, the output for punctuated transcript will take some minutes to complete (around 1 minute per 10-minute video)
58
+
4. Change the values in the second cell to include your URL etc.
59
+
5. Press CTRL+F9 or CMD+F9 to run the notebook.
53
60
54
61
## Run Locally 💻
55
62
56
63
I do not recommend running locally as it will download tensors and other stuff which are over 6gb. But if you want you can do this:
64
+
57
65
1. Clone the repository: `git clone https://github.com/therohitdas/Youtube-Transcript-Generator.git`
Punctuation model name can be taken from [here](https://huggingface.co/oliverguhr/fullstop-punctuation-multilang-large#languages).
116
129
117
130
Make sure to replace `https://www.youtube.com/watch?v=VIDEO_ID` with the actual URL of the YouTube video you want to process.
118
131
119
132
Feel free to copy and paste these examples into your terminal.
133
+
120
134
## Acknowledgments 🙌
121
135
122
136
This script utilizes the [youtube-transcript-api](https://github.com/jdepoix/youtube-transcript-api) and [fullstop-punctuation-multilang-large](https://huggingface.co/oliverguhr/fullstop-punctuation-multilang-large) libraries. Special thanks to their contributors.
123
137
124
138
Feel free to adapt and use the script based on your requirements. Enjoy the convenience of YouTube transcript processing!
0 commit comments