Editing NotebookLM


{{Draft}}

== Troubleshooting of NotebookLM ==
=== NotebookLM Workaround: Large Video to MP3 Conversion ===

'''Problem Description''': 

The original 2.3GB video file exceeded NotebookLM's strict 200MB limit for local uploads. As stated in their official documentation: "The current limit is 500,000 words per source or up to 200MB for local uploads."<ref>[https://support.google.com/notebooklm/answer/16269187?hl=en Frequently asked questions - NotebookLM Help]</ref> (每個檔案大小不得超過 200MB)

'''Solution:''' 

Use FFmpeg to convert the video to MP3 audio. Result: The file size was reduced to 126MB, successfully meeting the limit.

1. Download & install [https://www.ffmpeg.org/download.html Download FFmpeg]

2. Open the terminal

3. Command Used:

<pre>
ffmpeg -i input_video.mp4 -vn -c:a libmp3lame -q:a 2 output_audio.mp3
</pre>

Breakdown of flags:
* {{kbd | key=<nowiki>-i</nowiki>}} input_video.mp4: The input file.
* {{kbd | key=<nowiki>-vn</nowiki>}}: Skips the video stream (No Video).
* {{kbd | key=<nowiki>-c:a libmp3lame</nowiki>}}: Specifies the audio codec to use (LAME is standard for MP3).
* {{kbd | key=<nowiki>-q:a 2</nowiki>}}: Sets Variable Bit Rate (VBR) quality. 2 averages around 190 kbps (high quality). Use 0 for best quality or 4 for standard.
* {{kbd | key=<nowiki>output_audio.mp3</nowiki>}}: The name of the new file.


=== NotebookLM Error: Problem with Uploading YouTube URL ===

'''Problem Description''':

When attempting to upload YouTube URL to NotebookLM, users encounter the error message: "Something went wrong, Please try again later." 

'''Known Causes:'''

Based on user reports, this error typically occurs with:
* Videos that are too recent
* Non-public or private videos
* Content with access restrictions

'''Workarounds Solutions:'''

1. '''Convert video to transcript:'''
* Extract text content from video using [https://github.com/openai/whisper openai/whisper: Robust Speech Recognition via Large-Scale Weak Supervision] or othe Speech-To-Text tools
* Upload the resulting transcript file to NotebookLM

2. '''Convert video to images, then to PDF:'''
* Extract video frames as individual images
* Compile images into a single PDF document
* Upload the PDF to NotebookLM using [https://web-video-frame-toolkit.pages.dev/?lang=en Web Video Frame Toolkit] provides both conversion options mentioned above.

=== NotebookLM Error: Excel File Format Not Supported ===

'''Problem Description''':

When attempting to upload an Excel file directly to NotebookLM, users encounter an error due to unsupported file format.

'''Workaround Solution:'''

1. '''Upload via Google Drive:'''
* Upload the Excel file to Google Drive
* Return to NotebookLM and select Google Drive as the source
* Choose the previously uploaded Excel file

'''Known Limitations:'''

* Large files may still fail to import even via Google Drive
* Files of 4MB or larger are known to cause import issues

'''If Import Still Fails:'''

* Reduce the number of data rows
* Remove unnecessary columns to decrease file size


=== NotebookLM Error: SRT File Format Not Supported ===

'''Problem Description''':
When attempting to upload an SRT ((SubRip Subtitle file) directly to NotebookLM, users encounter an error due to unsupported file format.

'''Workaround Solution:''': Rename File Extension
# Locate the SRT file on your device
# Rename the file extension from .srt to .txt
# Upload the renamed .txt file directly to NotebookLM

'''Why This Works:'''
* SRT is a plain text format — changing the extension does not alter the content
* NotebookLM accepts .txt files, so renaming allows the file to pass format validation

== Further reading ==
* [https://support.google.com/notebooklm/answer/16215270?hl=en&co=GENIE.Platform%3DDesktop&sjid=7465465456853301270-NC#zippy Add or discover new sources for your notebook - Computer - NotebookLM Help]
* [https://support.google.com/notebooklm/?hl=en#topic=16164070 NotebookLM Help]

== References ==
<references />

[[Category: Tool]]
[[Category: Generative AI]]
[[Category: Video]]
[[Category: Audio]]
[[Category: Revised with LLMs]]