Releases: MatteoFasulo/Whisper-TikTok
Releases · MatteoFasulo/Whisper-TikTok
Whisper-TikTok v2.0.2
What's Changed
- Reddit2json addon by @duozokker in #30
- Add sidebar navigation and maximum words per line option by @MatteoFasulo in #33
- Merge reddit changes into dev branch for video generation feature by @MatteoFasulo in #34
- 31 webui reddit video generation by @MatteoFasulo in #36
- Reddit by @MatteoFasulo in #37
- Update README.md and arg_parser.py by @MatteoFasulo in #38
New Contributors
- @duozokker made their first contribution in #30
Full Changelog: v2.0.1...v2.0.2
Whisper-TikTok v2.0.1
What's Changed
- Essentially an adaptation of the new UI to the Readme by @YotomiY in #25
- Fix UTF-8 decoding in streamlit (solved by iPromis) by @YotomiY in #28
New Features
- Docker image associated with the repository for running the Web-UI using streamlit (locally). The public version is still available (without the Reddit feature which is still under development)
Full Changelog: v2.0.0...v2.0.1
Whisper-TikTok v2.0.0
UI
What's Changed
- Update TikTok Upload by @MatteoFasulo in #14
- Add stable-ts to requirements.txt and update video.json structure by @MatteoFasulo in #16
- Webui update: Your Backgrounds + Webui Stability Check + README.md Update by @YotomiY in #19
- Update absolute path of background folder and file name by @MatteoFasulo in #20
New Contributors
Full Changelog: v1.0.4...v2.0.0
Whisper-TikTok v1.0.4
New features:
Add subtitle format and font color as CLI args
--sub_format TEXT Subtitle format to use [u|i|b] (Default: u) | b (Bold), u (Underline), i (Italic)
--font_color TEXT Font color to use for subtitles in HEX format (Default: #FFFFFF). Suggested: #FFF000 (Yellow)
Fixed:
Full Changelog: v1.0.3...v1.0.4
Whisper-TikTok v1.0.3
What's Changed
- Update requirements.txt by @LucaTedeschini in #5
- 4 yt dlp corrupted download without metadata ffprobe by @MatteoFasulo in #6
New Contributors
- @LucaTedeschini made their first contribution in #5
- @MatteoFasulo made their first contribution in #6
Full Changelog: v1.0.2...v1.0.3
Whisper-TikTok v1.0.2
Updates
- Removed 'stable_whisper' dependency from project.
- Fixed path issues with 'background' folder
- word_options dict for customizing words highlighting as well as max line count and width
PR
Custom subtitle with HTML formatting submitted as PR
Full Changelog: v1.0.1...v1.0.2
Whisper-TikTok v1.0.1
What's Changed
- Added funding for project
- Added docs
- Auto-CI documentation files
- Added Issue template
- Added template for reporting a security vulnerability
- Added Pull Request template
Full Changelog: v1.0.0...v1.0.1
Whisper-TikTok v1.0.0
First complete release!
New features
- CLI support
- Support for all Whisper models (small and medium have the best trade-off betweeen accuracy and execution time)
- Support for non-english model with auto detect from mp3 file
- Support for random TTS voice with gender and language
- Better logging and console support with traceback
Fixed bugs
- Background folder #2
Full Changelog: v0.4...v1.0.0
Whisper-TikTok v0.4
- video.json instead of python dict
- add FFMPEG as a required package with guide on how to install it depending on your OS
- code refactoring
- removed dependency of CUDA with PyTorch. It will use the CPU if CUDA driver is not found.
Full Changelog: v0.3...v0.4
Whisper-TikTok v0.3
Demo
demo.mp4
Features
- fixed requirements.txt for CUDA PyTorch build
- more detailed documentation in readme
- changed repo name to Whisper-TikTok as ChatGPT has not been used yet
Full Changelog: v0.2...v0.3