YouTube to notes Convertor #27
Open
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- Added a new file `desktop.ini` to the `audio_extractor/samples` directory. - Created a new file `extractor.py` in the `audio_extractor` directory, which contains an abstract class `AudioExtractor` with an abstract method `extract_text`. - Added a new file `text_formatter.py` to the `text_formatter` directory, which contains a class `TextFormatter` with an `__init__` method and a `format_text` method. - Created a new file `img_handler.py` in the `slide_processor/extractors` directory, which contains a class `ImageHandler` with an `__init__` method and a `process_images` method. - Added a new file `text_cleaner.py` to the `text_formatter` directory, which contains a class `TextCleaner` with a static method `clean_text`. - Created a new file `markdown_formatter.py` in the `text_formatter` directory, which contains a class `MarkdownFormatter` with an `__init__` method, a `fix_formatting` method, and a `save_file` method. - Added a new file `fuzzy.py` to the `text/matching` directory, which contains a class `FuzzyMatcher` with a static method `fuzzy_matching`. - Created a new file `timestamp.py` in the `core/post_processing` directory, which contains a class `TimestampedNoteProcessor` with an `__init__` method and a `process_notes` method. - Added a new file `pdf_handler.py` to the `slide_processor/extractors` directory, which contains a function `extract_text_from_pdf` that extracts text from a PDF file using PyMuPDF. - Added a new file `pyproject.toml` to the root directory, which contains project metadata and dependencies.
Legedith
changed the title
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Tune AI PyCon India 2024 Contest
YouTube to notes Convertor
new : We can now generate and add Images to notes based off off sections. The prompts for these images are generated by a model hosted on Tune whereas the image gen is performed by a Tune Assistant (
tools/image/tune_img.py).This is done as a post processing step as not all note might need this. But when they do, the flow is:
(optional below)
note: this is a personal project, recently revamped using Tune. There's only so much you can do with the free context length of 20k when you have to process hours and hours of data. With Tune, I'm able to play around with models, change them on-the-go, and don't have to worry about the context length (as long as I have credits). In my testing, 16k tokens consumed around 2p of credits, which is pretty decent. If I process a 2 hour lecture, some 250k tokens in total, it would cost me around 3 rs. which is way better than the 3 days that I have to wait for Gemini to generate a decent set of notes for me.
Tune AI Products Used (Required)
Checklist
Demo
Refer these sample notes