Releases: octimot/StoryToolkitAI
StoryToolkitAI v0.23.2 - stories and speakers beta
What's new in this release
Main new features:
- Story Editor
make selections from your transcripts to create timelines
that can be imported into your editing or screenwriting software - Speaker Detection
detect speaker changes in audio either while ingesting or directly from the transcription window - Meta Transcription Segments
add speakers, notes, etc. to your transcripts using SHIFT+ENTER while editing a transcript - Major improvements to the Assistant
including the ability to provide new Transcriptions or Stories as responses
A few highlights:
- Export Story Editor content as EDL, XML or Fountain for editing in Resolve, Premiere, Final Cut Pro, Avid, etc.
- Ability to request transcription and story responses from the Assistant using the
[t]or[st]keywords - Ability to create new transcriptions, groups and stories based on Assistant responses
- Ability to change the models and settings used for the Assistant
- Ability to edit the Assistant chat history by enabling or disabling messages via right click context menu
- Ability to reuse Assistant prompts via right click context menu
- Ability to copy the Assistant prompts and conversations to clipboard via right click context menu
- Support for storytoolkit.ai Assistant Models
- Automatic Speaker Change Detection via Ingest or from the Transcription Window
- Ability to add transcript meta segments, such as speaker, notes, etc.
- Find and Replace for Transcriptions via Find window
- Improved handling of timecodes for EDL, XML, AVID DS and Fusion Comp exports
See CHANGELOG.md for more info.
Download links
This update is available early for Frequent Users only, to download see this post.
Everyone else can still download version 0.19.5 from here.
Help us make a better tool!
This tool and its code is free and open source.
If you find the tool useful, please consider becoming a patron.
This way, we can speed up development to work on cool features that you might need and keep up with rapid advancements in AI.
It also helps if you report any potential bugs or problems using the Issues tab here on Github: https://github.com/octimot/StoryToolkitAI/issues
Installation
macOS
Requirements
macOS 12.6 Monterey or newer
Intel or arm64 (M1, M2 etc.)
8 GB RAM
Installation Steps
Download the zip file that fits with your platform.
Unzip, copy the app file into your Applications folder, then double-click and it should start.
Please also read Info For All Platforms below.
Unsafe warnings
If you get something like "StoryToolkitAI is damaged and can't be opened", try to right click and then click open in the dropdown menu. It will warn you that it's unsafe, but that's just because we haven't notarized the tool with Apple.
If you still have problems opening it, try to take it out of quarantine: open up Terminal and execute:
xattr -d com.apple.quarantine ./StoryToolkitAI.0.23.1.arm64.app (or x86_64 if Intel).
More details on the issue here.
Windows
Requirements
Windows 10 or newer
Installation
Just follow the installer to get the tool on your machine.
Notes for Windows
Do not install this version of the tool in Program Files unless you know what you're doing. Some users have experienced permission issues which will make some features unusable.
If the tool returns errors or would not start after updating an older version, we recommend completely uninstalling and re-installing before reporting the issue.
If you get an error saying something like "Microsoft Defender SmartScreen prevented an unrecognized app from starting...", just press More info and Run anyway.
If the Speaker Detection process fails the first time you run it, it's most likely due to permission errors. So, close the app, right click and Run as administrator so that the tool can download the model into the cache folder. Once the first detection is performed, you should be able to run the tool with your normal user next time.
Also, see below:
Info for all platforms
You don't need to have Python installed on your machine for the standalone version to work, only if you want to use the Resolve API with the tool.
Download Permissions
If you get something like PermissionError: [Errno 13] Permission denied: '/Users/[your user]/.cache/torch/hub/trusted_list', it's most likely because the new standalone version uses different ssl certificates (issue #77) , so it's best if you delete the old hub cache, like this:
On macOS
Open terminal, and execute
rm -rf /Users/USERNAME/.cache/hub
rm -rf /Users/USERNAME/.cache/torch
rm -rf /Users/USERNAME/.cache/whisper
- replace USERNAME with your MacOS user
On Windows
Open CMD, and execute
rmdir /s /q C:\Users\USERNAME\.cache\hub
rmdir /s /q C:\Users\USERNAME\.cache\torch
rmdir /s /q C:\Users\USERNAME\.cache\whisper
- replace USERNAME with your Windows user
Keep in mind that if you do this, the first time you transcribe or search something it will need to re-download the models, so it will take a bit longer.
Make sure Resolve scripting is enabled
The tool only works with the Studio version of Davinci Resolve.
Davinci Resolve Preferences -> General, "External Scripting using", set to: Local
StoryToolkitAI v0.23.1 - stories and speakers alpha
What's new in this release
Main new features:
- Story Editor
make selections from your transcripts to create timelines
that can be imported into your editing or screenwriting software - Speaker Detection
detect speaker changes in audio either while ingesting or directly from the transcription window - Meta Transcription Segments
add speakers, notes, etc. to your transcripts using SHIFT+ENTER while editing a transcript - Major improvements to the Assistant
including the ability to provide new Transcriptions or Stories as responses
A few highlights:
- Export Story Editor content as EDL, XML or Fountain for editing in Resolve, Premiere, Final Cut Pro, Avid, etc.
- Ability to request transcription and story responses from the Assistant using the
[t]or[st]keywords - Ability to create new transcriptions, groups and stories based on Assistant responses
- Ability to change the models and settings used for the Assistant
- Ability to edit the Assistant chat history by enabling or disabling messages via right click context menu
- Ability to reuse Assistant prompts via right click context menu
- Ability to copy the Assistant prompts and conversations to clipboard via right click context menu
- Support for storytoolkit.ai Assistant Models
- Automatic Speaker Change Detection via Ingest or from the Transcription Window
- Ability to add transcript meta segments, such as speaker, notes, etc.
- Find and Replace for Transcriptions via Find window
- Improved handling of timecodes for EDL, XML, AVID DS and Fusion Comp exports
Download links
This update is available early for Frequent Users only, to download see this post.
Everyone else can still download version 0.19.5 from here.
Help us make a better tool!
This tool and its code is free and open source.
If you find the tool useful, please consider becoming a patron.
This way, we can speed up development to work on cool features that you might need and keep up with rapid advancements in AI.
It also helps if you report any potential bugs or problems using the Issues tab here on Github: https://github.com/octimot/StoryToolkitAI/issues
Installation
macOS
Requirements
macOS 12.6 Monterey or newer
Intel or arm64 (M1, M2 etc.)
8 GB RAM
Installation Steps
Download the zip file that fits with your platform.
Unzip, copy the app file into your Applications folder, then double-click and it should start.
Please also read Info For All Platforms below.
Unsafe warnings
If you get something like "StoryToolkitAI is damaged and can't be opened", try to right click and then click open in the dropdown menu. It will warn you that it's unsafe, but that's just because we haven't notarized the tool with Apple.
If you still have problems opening it, try to take it out of quarantine: open up Terminal and execute:
xattr -d com.apple.quarantine ./StoryToolkitAI.0.23.1.arm64.app (or x86_64 if Intel).
More details on the issue here.
Windows
Requirements
Windows 10 or newer
Installation
Just follow the installer to get the tool on your machine.
Notes for Windows
Do not install this version of the tool in Program Files unless you know what you're doing. Some users have experienced permission issues which will make some features unusable.
If the tool returns errors or would not start after updating an older version, we recommend completely uninstalling and re-installing before reporting the issue.
If you get an error saying something like "Microsoft Defender SmartScreen prevented an unrecognized app from starting...", just press More info and Run anyway.
If the Speaker Detection process fails the first time you run it, it's most likely due to permission errors. So, close the app, right click and Run as administrator so that the tool can download the model into the cache folder. Once the first detection is performed, you should be able to run the tool with your normal user next time.
Also, see below:
Info for all platforms
You don't need to have Python installed on your machine for the standalone version to work, only if you want to use the Resolve API with the tool.
Download Permissions
If you get something like PermissionError: [Errno 13] Permission denied: '/Users/[your user]/.cache/torch/hub/trusted_list', it's most likely because the new standalone version uses different ssl certificates (issue #77) , so it's best if you delete the old hub cache, like this:
On macOS
Open terminal, and execute
rm -rf /Users/USERNAME/.cache/hub
rm -rf /Users/USERNAME/.cache/torch
rm -rf /Users/USERNAME/.cache/whisper
- replace USERNAME with your MacOS user
On Windows
Open CMD, and execute
rmdir /s /q C:\Users\USERNAME\.cache\hub
rmdir /s /q C:\Users\USERNAME\.cache\torch
rmdir /s /q C:\Users\USERNAME\.cache\whisper
- replace USERNAME with your Windows user
Keep in mind that if you do this, the first time you transcribe or search something it will need to re-download the models, so it will take a bit longer.
Make sure Resolve scripting is enabled
The tool only works with the Studio version of Davinci Resolve.
Davinci Resolve Preferences -> General, "External Scripting using", set to: Local
StoryToolkitAI v0.19.5 - alpha video search public
Release Info
This release focuses on improving some of the functions introduced in the previous versions.
Key Features
- Full video indexing and search (v. 0.19.2+ - How-To)
- Free Automatic Transcriptions on your local machine
- Free Automatic Translation to English on your local machine
- ChatGPT integration - talk to AI about your content, or generate new ideas
- Search Content intuitively without having to type in exact words
- Transcript Groups - group transcript lines into whatever you need to find them easier
- Automatic Question detection in transcripts
- Multi-format export of transcripts, including SRT, TXT, AVID DS and as Fusion Text node
- Import of existing SRT files
- Easy copy of timecoded transcript text to clipboard etc.
Resolve Studio Integrations
- Mark and Navigate Resolve Timelines via Transcript, plus other handy Resolve-only features
- Advanced Search of Resolve timeline markers using AI
- Copy Resolve timeline markers to transcript and vice-versa for advanced search
- Direct import of subtitles into Resolve bin
For a detalied list of version to version changes, see the Changelog.
Please report any issues that you might encounter here.
StoryToolkitAI Patreon
If you find the tool useful, please consider becoming a patron.
This way, we can speed up development to work on cool features that you might need and keep up with rapid advancements in AI.
Installation:
IMPORTANT NOTE
We recommend backing up the configuration folder your user directory if you update from versions earlier than 0.18.0, since this version contains a major overhaul of the tool file structure.
macOS
Requirements
macOS 12.6 Monterey or newer
Intel or arm64 (M1, M2 etc.)
8 GB RAM
Installation Steps
Download the zip file that fits with your platform.
Unzip, copy the app file into your Applications folder, then double-click and it should start.
Please also read Info For All Platforms below.
Unsafe warnings
If you get something like "StoryToolkitAI is damaged and can't be opened", try to right click and then click open in the dropdown menu. It will warn you that it's unsafe, but that's just because we haven't notarized the tool with Apple.
If you still have problems opening it, try to take it out of quarantine: open up Terminal and execute:
xattr -d com.apple.quarantine ./StoryToolkitAI.0.18.0.1.arm64.app (or x86_64 if Intel).
More details on the issue here.
Console Mode
Note: The standalone version does not open up the console. If you're interested to see what is happening behind the hood, open a terminal in the folder where you have the tool (typically in your Downloads or Applications folder) and run this command:
./[STORYTOOLKITAI_APP_NAME].app/Contents/MacOS/[STORYTOOLKIT_APP_NAME]
# (but replace [STORYTOOLKIT_APP_NAME] with the name of your release)
Windows
Requirements
Windows 10 or newer
CUDA-enabled graphics card
Installation
Just follow the installer to get the tool on your machine.
Notes for Windows
Do not install this version of the tool in Program Files unless you know what you're doing. Some users have experienced permission issues which will make some features unusable.
If the tool returns errors or would not start after updating an older version, we recommend completely uninstalling and re-installing before reporting the issue.
You don't need to have Python installed on your machine for the standalone version to work, only if you want to use the Resolve API with the tool.
But, if you have Resolve Studio installed and the tool doesn't start, it's probably because you have multiple Python versions installed on your machine (see #22). This is because Resolve API seems to choose the newest version of Python that is available on the machine, while the tool uses Python 3.10. If you're feeling brave, we recommend installing directly via git and creating an appropriate virtual environment that matches the one that Resolve is using.
Also, if you do need Python for other reasons, make sure you install it from the official Python website and only use other Python installers / versions if you know what you're doing.
If you see an error mentioning python_check, start the tool with the --skip-python-check argument.
Also, see below:
Info for all platforms
Download Permissions
If you get something like PermissionError: [Errno 13] Permission denied: '/Users/[your user]/.cache/torch/hub/trusted_list', it's most likely because the new standalone version uses different ssl certificates (issue #77) , so it's best if you delete the old hub cache, like this:
On macOS
Open terminal, and execute
rm -rf /Users/USERNAME/.cache/hub
rm -rf /Users/USERNAME/.cache/torch
rm -rf /Users/USERNAME/.cache/whisper
- replace USERNAME with your MacOS user
On Windows
Open CMD, and execute
rmdir /s /q C:\Users\USERNAME\.cache\hub
rmdir /s /q C:\Users\USERNAME\.cache\torch
rmdir /s /q C:\Users\USERNAME\.cache\whisper
- replace USERNAME with your Windows user
Keep in mind that if you do this, the first time you transcribe or search something it will need to re-download the models, so it will take a bit longer.
Make sure Resolve scripting is enabled
The tool only works with the Studio version of Davinci Resolve.
Davinci Resolve Preferences -> General, "External Scripting using", set to: Local
transcription_WAV
For the optimum transcription quality, try to create the transcription_WAV render preset in Resolve following this info, or simply use WAV audio files or any other format that uses Linear PCM for audio when transcribing - preferably mono.
Downloading models to the cache
The first time you transcribe something with a specific model, it will take a bit longer to start the process because Whisper needs to download the model file (around 1.5GB for the medium model) on your local machine. But, after the model is saved on your machine, transcriptions will take less. The same applies to the advanced search function.
StoryToolkitAI v0.19.4 - alpha video search
What's new in this release
AI-Powered Text to Image Search in Videos - find anything you can describe with words in your footage (quick demo)!
And a new UI that makes the tool look less like it's supposed to run on a Windows 95 machine...
A few highlights:
- Advanced Video Search - find anything you can describe with words in your footage
- New User Interface
- Ingesting of video content which includes AI based scene detection and video content indexing
- New Ingest window to handle both Transcriptions and Video Indexing
- A better Queue system which allows queuing and managing of all CPU/GPU intensive tasks
- Improved Advanced Search
- Easier handling of multiple files when using the Transcription Settings window
- Right-click context menu on transcripts for easier access to functions
- Improved handling of Transcript Groups
- Significant code refactoring and cleanup, multiple bug fixes etc.
Download links
This update is available early for Frequent User and Producer tiers only, to download see this post.
For everyone else, download version 0.17.16 from here.
StoryToolkitAI Patreon
BTW, if you find the tool useful, please consider becoming a patron.
This way, we can speed up development to work on cool features that you might need and keep up with rapid advancements in AI.
Installation:
IMPORTANT NOTE
We recommend backing up the configuration folder your user directory if you update from versions earlier than 0.18.0, since this version contains a major overhaul of the tool file structure.
macOS
Requirements
macOS 12.6 Monterey or newer
Intel or arm64 (M1, M2 etc.)
8 GB RAM
Installation Steps
Download the zip file that fits with your platform.
Unzip, copy the app file into your Applications folder, then double-click and it should start.
Please also read Info For All Platforms below.
Unsafe warnings
If you get something like "StoryToolkitAI is damaged and can't be opened", try to right click and then click open in the dropdown menu. It will warn you that it's unsafe, but that's just because we haven't notarized the tool with Apple.
If you still have problems opening it, try to take it out of quarantine: open up Terminal and execute:
xattr -d com.apple.quarantine ./StoryToolkitAI.0.18.0.1.arm64.app (or x86_64 if Intel).
More details on the issue here.
Console Mode
Note: The standalone version does not open up the console. If you're interested to see what is happening behind the hood, open a terminal in the folder where you have the tool (typically in your Downloads or Applications folder) and run this command:
./[STORYTOOLKITAI_APP_NAME].app/Contents/MacOS/[STORYTOOLKIT_APP_NAME]
# (but replace [STORYTOOLKIT_APP_NAME] with the name of your release)
Windows
Requirements
Windows 10 or newer
CUDA-enabled graphics card
Installation
Just follow the installer to get the tool on your machine.
Notes for Windows
Do not install this version of the tool in Program Files unless you know what you're doing. Some users have experienced permission issues which will make some features unusable.
In case some functions return errors, a first good step is to uninstall the tool using the uninstaller and then re-install it again.
You don't need to have Python installed on your machine for the standalone version to work, only if you want to use the Resolve API with the tool.
But, if you have Resolve Studio installed and the tool doesn't start, it's probably because you have multiple Python versions installed on your machine (see #22). This is because Resolve API seems to choose the newest version of Python that is available on the machine, while the tool uses Python 3.10. If you're feeling brave, we recommend installing directly via git and creating an appropriate virtual environment that matches the one that Resolve is using.
Also, if you do need Python for other reasons, make sure you install it from the official Python website and only use other Python installers / versions if you know what you're doing.
If you see an error mentioning python_check, start the tool with the --skip-python-check argument.
Also, see below:
Info for all platforms
Download Permissions
If you get something like PermissionError: [Errno 13] Permission denied: '/Users/[your user]/.cache/torch/hub/trusted_list', it's most likely because the new standalone version uses different ssl certificates (issue #77) , so it's best if you delete the old hub cache, like this:
On macOS
Open terminal, and execute
rm -rf /Users/USERNAME/.cache/hub
rm -rf /Users/USERNAME/.cache/torch
rm -rf /Users/USERNAME/.cache/whisper
- replace USERNAME with your MacOS user
On Windows
Open CMD, and execute
rmdir /s /q C:\Users\USERNAME\.cache\hub
rmdir /s /q C:\Users\USERNAME\.cache\torch
rmdir /s /q C:\Users\USERNAME\.cache\whisper
- replace USERNAME with your Windows user
Keep in mind that if you do this, the first time you transcribe or search something it will need to re-download the models, so it will take a bit longer.
Make sure Resolve scripting is enabled
The tool only works with the Studio version of Davinci Resolve.
Davinci Resolve Preferences -> General, "External Scripting using", set to: Local
transcription_WAV
For the optimum transcription quality, try to create the transcription_WAV render preset in Resolve following this info, or simply use WAV audio files or any other format that uses Linear PCM for audio when transcribing - preferably mono.
Downloading models to the cache
The first time you transcribe something with a specific model, it will take a bit longer to start the process because Whisper needs to download the model file (around 1.5GB for the medium model) on your local machine. But, after the model is saved on your machine, transcriptions will take less. The same applies to the advanced search function.
StoryToolkitAI v0.18.3 - alpha rain
What's new in this release
Timecodes, AVID DS export, Blackmagic Fusion Text+ and the Group Questions with AI feature
A few highlights:
- Group Questions - use AI to automatically detect and group questions from transcripts
- Ability to export transcript to Blackmagic Fusion Text+ via menu button
- Ability to export to Avid DS format via menu button
- Time Intervals in the Transcription Settings Window can now be set using timecodes too
- Ability to use timecodes for copying transcript segments even without Resolve API connection
- fixes, optimizations etc.
For a detailed list of changes, see CHANGELOG.
Download links
This update is available early for Frequent User and Producer tiers only, to download see this post.
For everyone else, download version 0.17.16 from here.
StoryToolkitAI Patreon
BTW, if you find the tool useful, please consider becoming a patron.
This way, we can speed up development to work on cool features that you might need and keep up with rapid advancements in AI.
Installation:
IMPORTANT NOTE
We recommend backing up the configuration folder your user directory if you update from versions earlier than 0.18.0, since this version contains a major overhaul of the tool file structure.
macOS
Requirements
macOS 12.6 Monterey or newer
Intel or arm64 (M1, M2 etc.)
8 GB RAM
Installation Steps
Download the zip file that fits with your platform.
Unzip, copy the app file into your Applications folder, then double-click and it should start.
Please also read Info For All Platforms below.
Unsafe warnings
If you get something like "StoryToolkitAI is damaged and can't be opened", try to right click and then click open in the dropdown menu. It will warn you that it's unsafe, but that's just because we haven't notarized the tool with Apple.
If you still have problems opening it, try to take it out of quarantine: open up Terminal and execute:
xattr -d com.apple.quarantine ./StoryToolkitAI.0.18.0.1.arm64.app (or x86_64 if Intel).
More details on the issue here.
Console Mode
Note: The standalone version does not open up the console. If you're interested to see what is happening behind the hood, open a terminal in the folder where you have the tool (typically in your Downloads or Applications folder) and run this command:
./[STORYTOOLKITAI_APP_NAME].app/Contents/MacOS/[STORYTOOLKIT_APP_NAME]
# (but replace [STORYTOOLKIT_APP_NAME] with the name of your release)
Windows
Requirements
Windows 10 or newer
CUDA-enabled graphics card
Installation
Just follow the installer to get the tool on your machine.
Notes for Windows
Do not install this version of the tool in Program Files unless you know what you're doing. Some users have experienced permission issues which will make some features unusable.
In case some functions return errors, a first good step is to uninstall the tool using the uninstaller and then re-install it again.
You don't need to have Python installed on your machine for the standalone version to work.
But, if you have Resolve Studio installed and the tool doesn't start, it's probably because you have multiple Python versions installed on your machine (see #22). This is because Resolve API seems to choose the newest version of Python that is available on the machine, while the tool uses 3.9.13. If you're feeling brave, we recommend installing directly via git and creating an appropriate virtual environment that matches the one that Resolve is using.
Also, if you do need Python for other reasons, make sure you install it from the official Python website and only use other Python installers / versions if you know what you're doing.
If you see an error mentioning python_check, start the tool with the --skip-python-check argument.
Also, see below:
Info for all platforms
Download Permissions
If you get something like PermissionError: [Errno 13] Permission denied: '/Users/[your user]/.cache/torch/hub/trusted_list', it's most likely because the new standalone version uses different ssl certificates (issue #77) , so it's best if you delete the old hub cache, like this:
On macOS
Open terminal, and execute
rm -rf /Users/USERNAME/.cache/hub
rm -rf /Users/USERNAME/.cache/torch
rm -rf /Users/USERNAME/.cache/whisper
- replace USERNAME with your MacOS user
On Windows
Open CMD, and execute
rmdir /s /q C:\Users\USERNAME\.cache\hub
rmdir /s /q C:\Users\USERNAME\.cache\torch
rmdir /s /q C:\Users\USERNAME\.cache\whisper
- replace USERNAME with your Windows user
Keep in mind that if you do this, the first time you transcribe or search something it will need to re-download the models, so it will take a bit longer.
Make sure Resolve scripting is enabled
The tool only works with the Studio version of Davinci Resolve.
Davinci Resolve Preferences -> General, "External Scripting using", set to: Local
transcription_WAV
For the optimum transcription quality, try to create the transcription_WAV render preset in Resolve following this info, or simply use WAV audio files or any other format that uses Linear PCM for audio when transcribing - preferably mono.
Downloading models to the cache
The first time you transcribe something with a specific model, it will take a bit longer to start the process because Whisper needs to download the model file (around 1.5GB for the medium model) on your local machine. But, after the model is saved on your machine, transcriptions will take less. The same applies to the advanced search function.
StoryToolkitAI v0.18.1 - still in alpha
What's new in this release
We're slowly pushing updates that include access to ChatGPT from the tool, to have a state-of-the-art AI by your side to look into transcripts, markers, projects etc. and do all sorts of interesting stuff in the future, from transcript summarization, to grouping content and maybe even coding timelines to re-import into your editing suite - more tips and ideas regarding this integration soon via Patreon!
A few highlights:
- Improved usability of Transcription Window (it still looks like we're in 2005, but more UI updates to come!)
- Ability to export transcripts as SRT and Text files via menu buttons and CMD/CTRL+SHIFT+S ( #84)
- Most transcription related features are now available via menu buttons
- Custom Punctuation Marks via Preferences Window to be used for splitting segments on punctuation (Patron request)
- Warning when trying to add markers to non-linked timeline in Resolve
- Additional menu shortcuts
- Fixed some bad bad bugs which prevented some Resolve API commands
For a detailed list of changes, see CHANGELOG.
Early Update Access for Patreon Frequent Users and Producers
This update is available early for Frequent User and Producer tiers only, see this post on Patreon
Version 0.17.16 is still available publicly for everyone else here.
StoryToolkitAI Patreon
BTW, if you find the tool useful, please consider becoming a patron.
This way, we can speed up development to work on cool features that you might need and keep up with rapid advancements in AI.
Installation:
IMPORTANT NOTE
We recommend backing up the configuration folder your user directory if you update from versions earlier than 0.18.0, since this version contains a major overhaul of the tool file structure.
macOS
Requirements
macOS 12.6 Monterey or newer
Intel or arm64 (M1, M2 etc.)
8 GB RAM
Installation Steps
Download the zip file that fits with your platform.
Unzip, copy the app file into your Applications folder, then double-click and it should start.
Please also read Info For All Platforms below.
Unsafe warnings
If you get something like "StoryToolkitAI is damaged and can't be opened", try to right click and then click open in the dropdown menu. It will warn you that it's unsafe, but that's just because we haven't notarized the tool with Apple.
If you still have problems opening it, try to take it out of quarantine: open up Terminal and execute:
xattr -d com.apple.quarantine ./StoryToolkitAI.0.18.0.1.arm64.app (or x86_64 if Intel).
More details on the issue here.
Console Mode
Note: The standalone version does not open up the console. If you're interested to see what is happening behind the hood, open a terminal in the folder where you have the tool (typically in your Downloads or Applications folder) and run this command:
./[STORYTOOLKITAI_APP_NAME].app/Contents/MacOS/[STORYTOOLKIT_APP_NAME]
# (but replace [STORYTOOLKIT_APP_NAME] with the name of your release)
Windows
Requirements
Windows 10 or newer
CUDA-enabled graphics card
Installation
Just follow the installer to get the tool on your machine.
Notes for Windows
Do not install this version of the tool in Program Files unless you know what you're doing. Some users have experienced permission issues which will make some features unusable.
In case some functions return errors, a first good step is to uninstall the tool using the uninstaller and then re-install it again.
You don't need to have Python installed on your machine for the standalone version to work.
But, if you have Resolve Studio installed and the tool doesn't start, it's probably because you have multiple Python versions installed on your machine (see #22). This is because Resolve API seems to choose the newest version of Python that is available on the machine, while the tool uses 3.9.13. If you're feeling brave, we recommend installing directly via git and creating an appropriate virtual environment that matches the one that Resolve is using.
Also, if you do need Python for other reasons, make sure you install it from the official Python website and only use other Python installers / versions if you know what you're doing.
If you see an error mentioning python_check, start the tool with the --skip-python-check argument.
Also, see below:
Info for all platforms
Download Permissions
If you get something like PermissionError: [Errno 13] Permission denied: '/Users/[your user]/.cache/torch/hub/trusted_list', it's most likely because the new standalone version uses different ssl certificates (issue #77) , so it's best if you delete the old hub cache, like this:
On macOS
Open terminal, and execute
rm -rf /Users/USERNAME/.cache/hub
rm -rf /Users/USERNAME/.cache/torch
rm -rf /Users/USERNAME/.cache/whisper
- replace USERNAME with your MacOS user
On Windows
Open CMD, and execute
rmdir /s /q C:\Users\USERNAME\.cache\hub
rmdir /s /q C:\Users\USERNAME\.cache\torch
rmdir /s /q C:\Users\USERNAME\.cache\whisper
- replace USERNAME with your Windows user
Keep in mind that if you do this, the first time you transcribe or search something it will need to re-download the models, so it will take a bit longer.
Make sure Resolve scripting is enabled
The tool only works with the Studio version of Davinci Resolve.
Davinci Resolve Preferences -> General, "External Scripting using", set to: Local
transcription_WAV
For the optimum transcription quality, try to create the transcription_WAV render preset in Resolve following this info, or simply use WAV audio files or any other format that uses Linear PCM for audio when transcribing - preferably mono.
Downloading models to the cache
The first time you transcribe something with a specific model, it will take a bit longer to start the process because Whisper needs to download the model file (around 1.5GB for the medium model) on your local machine. But, after the model is saved on your machine, transcriptions will take less. The same applies to the advanced search function.
StoryToolkitAI v0.18.0 - almost not alpha
What's new in this release
We're slowly pushing updates that include access to ChatGPT from the tool, to have a state-of-the-art AI by your side to look into transcripts, markers, projects etc. and do all sorts of interesting stuff in the future, from transcript summarization, to grouping content and maybe even coding timelines to re-import into your editing suite - more tips and ideas regarding this integration soon via Patreon!
A few highlights:
- Direct interface to ChatGPT via Assistant window (see above)
- Ability to split transcript lines using Max. Words Per Line and Max. Characters Per Line and Split On Punctuation settings (see #42)
- Ability to prevent annoying gaps between transcript lines and resulting subtitles (see #88)
- Ability to cancel transcriptions from the queue (#35)
- Additional shortcuts and improvements to Transcription Groups
- Additional menu shortcuts
- Improved handling of Resolve API, including menu buttons to Connect to- or Disable API (see #86)
For a detailed list of changes, see CHANGELOG.
Early Update Access for Patreon Frequent Users and Producers
This update is available early for Frequent User and Producer tiers only, please see this post.
Version 0.17.16 is still available publicly for everyone else here.
StoryToolkitAI Patreon
BTW, if you find the tool useful, please consider becoming a patron.
This way, we can speed up development to work on cool features that you might need and keep up with rapid advancements in AI.
Installation:
IMPORTANT NOTE
We recommend backing up the configuration folder your user directory since this version contains a major overhaul of the tool file structure.
macOS
Requirements
macOS 12.6 Monterey or newer
Intel or arm64 (M1, M2 etc.)
8 GB RAM
Installation Steps
Download the zip file that fits with your platform.
Unzip, copy the app file into your Applications folder, then double-click and it should start.
Please also read Info For All Platforms below.
Unsafe warnings
If you get something like "StoryToolkitAI is damaged and can't be opened", try to right click and then click open in the dropdown menu. It will warn you that it's unsafe, but that's just because we haven't notarized the tool with Apple.
If you still have problems opening it, try to take it out of quarantine: open up Terminal and execute:
xattr -d com.apple.quarantine ./StoryToolkitAI.0.18.0.1.arm64.app (or x86_64 if Intel).
More details on the issue here.
Console Mode
Note: The standalone version does not open up the console. If you're interested to see what is happening behind the hood, open a terminal in the folder where you have the tool (typically in your Downloads or Applications folder) and run this command:
./[STORYTOOLKITAI_APP_NAME].app/Contents/MacOS/[STORYTOOLKIT_APP_NAME]
# (but replace [STORYTOOLKIT_APP_NAME] with the name of your release)
Windows
Requirements
Windows 10 or newer
CUDA-enabled graphics card
Installation
Just follow the installer to get the tool on your machine.
Notes for Windows
Do not install this version of the tool in Program Files unless you know what you're doing. Some users have experienced permission issues which will make some features unusable.
In case some functions return errors, a first good step is to uninstall the tool using the uninstaller and then re-install it again.
You don't need to have Python installed on your machine for the standalone version to work.
But, if you have Resolve Studio installed and the tool doesn't start, it's probably because you have multiple Python versions installed on your machine (see #22). This is because Resolve API seems to choose the newest version of Python that is available on the machine, while the tool uses 3.9.13. If you're feeling brave, we recommend installing directly via git and creating an appropriate virtual environment that matches the one that Resolve is using.
Also, if you do need Python for other reasons, make sure you install it from the official Python website and only use other Python installers / versions if you know what you're doing.
If you see an error mentioning python_check, start the tool with the --skip-python-check argument.
Also, see below:
Info for all platforms
Download Permissions
If you get something like PermissionError: [Errno 13] Permission denied: '/Users/[your user]/.cache/torch/hub/trusted_list', it's most likely because the new standalone version uses different ssl certificates (issue #77) , so it's best if you delete the old hub cache, like this:
On macOS
Open terminal, and execute
rm -rf /Users/USERNAME/.cache/hub
rm -rf /Users/USERNAME/.cache/torch
rm -rf /Users/USERNAME/.cache/whisper
- replace USERNAME with your MacOS user
On Windows
Open CMD, and execute
rmdir /s /q C:\Users\USERNAME\.cache\hub
rmdir /s /q C:\Users\USERNAME\.cache\torch
rmdir /s /q C:\Users\USERNAME\.cache\whisper
- replace USERNAME with your Windows user
Keep in mind that if you do this, the first time you transcribe or search something it will need to re-download the models, so it will take a bit longer.
Make sure Resolve scripting is enabled
The tool only works with the Studio version of Davinci Resolve.
Davinci Resolve Preferences -> General, "External Scripting using", set to: Local
transcription_WAV
For the optimum transcription quality, try to create the transcription_WAV render preset in Resolve following this info, or simply use WAV audio files or any other format that uses Linear PCM for audio when transcribing - preferably mono.
Downloading models to the cache
The first time you transcribe something with a specific model, it will take a bit longer to start the process because Whisper needs to download the model file (around 1.5GB for the medium model) on your local machine. But, after the model is saved on your machine, transcriptions will take less. The same applies to the advanced search function.
StoryToolkitAI v0.17.16 - extra alpha
Download the standalone release here:
StoryToolkitAI 0.17.16.2 for Mac M1, M2 etc.
StoryToolkitAI 0.17.16 for Mac Intel
StoryToolkitAI 0.17.16.2 for Windows (only for CUDA GPUs)
Please report if something doesn't work as expected. So far, only a handful of users have tested this new standalone version!
StoryToolkitAI Patreon
If you find the tool useful, please consider becoming a patron.
This way, we can speed up development to work on cool features that you might need and keep up with rapid advancements in AI.
What's new in this release
Many things have been added and optimized, here are a few highlights:
- Increased Time Precision improves transcription timings (more updates on this soon)
- Pre-detect speech to reduce hallucinations on silence and improve speed
- Support for Whisper large-v2 model
- Transcription progress now shown in the Transcription Log window
- Significantly faster Advanced Search when dealing with many transcriptions or files
- Advanced Search on text files and Resolve markers (via project.json)
- CMD/CTRL+M in the Transcription Window selects all segments between markers filtered by color or name from Resolve timeline
- SHIFT+A if text is selected in the Transcription Window will now select all segments under that text
- Better Find function in Transcription and Advanced Search windows
- Better handling of Resolve markers
- many UI and backend optimizations and improvements
For a detailed list of changes since version 0.17.7, see CHANGELOG.
Installation:
macOS
Requirements
macOS 12.6 Monterey or newer
Intel or arm64 (M1, M2 etc.)
8 GB RAM
Installation Steps
Download the zip file that fits with your platform.
Unzip, copy the app file into your Applications folder, then double-click and it should start.
Please also read Info For All Platforms below.
Unsafe warnings
If you get something like "StoryToolkitAI is damaged and can't be opened", try to right click and then click open in the dropdown menu. It will warn you that it's unsafe, but that's just because we haven't notarized the tool with Apple.
If you still have problems opening it, try to take it out of quarantine: open up Terminal and execute:
xattr -d com.apple.quarantine ./StoryToolkitAI.0.17.16.arm64.app (or x86_64 if Intel).
More details on the issue here.
Console Mode
Note: The standalone version does not open up the console. If you're interested to see what is happening behind the hood, open a terminal in the folder where you have the tool (typically in your Downloads or Applications folder) and run this command:
./[STORYTOOLKITAI_APP_NAME].app/Contents/MacOS/[STORYTOOLKIT_APP_NAME]
# (but replace [STORYTOOLKIT_APP_NAME] with the name of your release)
Windows
Requirements
Windows 10 or newer
CUDA-enabled graphics card
Easy Installation
Just follow the installer to get the tool on your machine.
Notes for Windows
Do not install this version of the tool in Program Files unless you know what you're doing. Some users have experienced permission issues which will make some features unusable.
You don't need to have Python installed on your machine for the standalone version to work.
But, if you have Resolve Studio installed and the tool doesn't start, it's probably because you have multiple Python versions installed on your machine (see #22). This is because Resolve API seems to choose the newest version of Python that is available on the machine, while the tool uses 3.9.13. If you're feeling brave, we recommend installing directly via git and creating an appropriate virtual environment that matches the one that Resolve is using.
Also, if you do need Python for other reasons, make sure you install it from the official Python website and only use other Python installers / versions if you know what you're doing.
If you see an error mentioning python_check, start the tool with the --skip-python-check argument.
Also, see below:
Info for all platforms
Download Permissions
If you get something like PermissionError: [Errno 13] Permission denied: '/Users/[your user]/.cache/torch/hub/trusted_list', it's most likely because the new standalone version uses different ssl certificates (issue #77) , so it's best if you delete the old hub cache, like this:
On macOS
Open terminal, and execute
rm -rf /Users/USERNAME/.cache/hub
rm -rf /Users/USERNAME/.cache/torch
rm -rf /Users/USERNAME/.cache/whisper
- replace USERNAME with your MacOS user
On Windows
Open CMD, and execute
rmdir /s /q C:\Users\USERNAME\.cache\hub
rmdir /s /q C:\Users\USERNAME\.cache\torch
rmdir /s /q C:\Users\USERNAME\.cache\whisper
- replace USERNAME with your Windows user
Keep in mind that if you do this, the first time you transcribe or search something it will need to re-download the models, so it will take a bit longer.
Make sure Resolve scripting is enabled
The tool only works with the Studio version of Davinci Resolve.
Davinci Resolve Preferences -> General, "External Scripting using", set to: Local
transcription_WAV
For the optimum transcription quality, try to create the transcription_WAV render preset in Resolve following this info, or simply use WAV audio files or any other format that uses Linear PCM for audio when transcribing - preferably mono.
Downloading models to the cache
The first time you transcribe something with a specific model, it will take a bit longer to start the process because Whisper needs to download the model file (around 1.5GB for the medium model) on your local machine. But, after the model is saved on your machine, transcriptions will take less. The same applies to the advanced search function.
StoryToolkitAI v0.17.5 - ultra alpha
Download the standalone release here:
StoryToolkitAI 0.17.5 for Mac M1
StoryToolkitAI 0.17.5 for Mac Intel
StoryToolkitAI 0.17.5 for Windows (only for CUDA GPUs)
Please report if something doesn't work as expected.
Info
The standalone version is way behind the git version of StoryToolkitAI. This is because compiling the app into Windows and Mac executables isn't really a walk in the park. We're working on getting the new standalone version up to date within a few days.
However, if you find the tool useful, please consider supporting the development, and you may enjoy some perks while doing so. If not, no worries, the tool will continue to be open and free for as long as we can support it!
What's new in this release
- Transcript Groups allows you to group transcript segments and recall them later
(CMD/CTRL+G to add segment to group, then use SHIFT+G in the transcription window to open up the transcript group window.) - Copy transcript lines with timecode using CMD/CTRL+SHIFT+C (then simply paste them wherever you want)
- New transcript editing shortcuts: CMD/CTRL+A selects all transcript segments, BACKSPACE deletes active segment
- Improved Resolve API communication and logging
- When transcribing Resolve timelines, the timeline name, frame rate and start timecode is saved to the transcription.json file
- Other minor fixes, including transcript scrollbar, more detailed logging etc.
Installation:
If you never installed the tool on your machine, please read this. Otherwise, simply replacing the old app will work.
IMPORTANT
The following steps describe how you should install ffmpeg on your machine to benefit from the full functionality of StoryToolkitAI. In reality, you do not need to have ffmpeg installed unless you're transcribing files that don't use the Linear PCM codec for their audio. As a matter of fact, other formats work without ffmpeg aswel, but we just haven't tested it widely enough to know for sure which ones.
So, before torturing yourself with the following steps (unless you like the terminal or light pain in general), try to create the transcription_WAV render preset in Resolve following this info, or simply use WAV audio files or any other format that uses Linear PCM for audio when transcribing, and if it's good enough for you just ignore the following steps (except the permissions part on MacOS, if you get warnings when you start the app). Also, make sure that you're using mono audio to prevent the algorithm from ignoring some of the channels.
Obviously if you're using the tool for other functions apart from transcriptions, you absolutely do not need ffmpeg on your machine.
If this works for enough people we'll take out the FFMPEG warning at the start of the app and just provide these steps for advanced use cases. What do you think?
Mac OS
You need to install ffmpeg manually from the terminal and the easiest way to do that is via Homebrew:
1. Install Homebrew
Open up a Terminal window, then run:
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
If Homebrew installation fails, first make sure you have xcode installed:
xcode-select --install
Note: if Homebrew installation fails for any reason, a good starting point is this page.
2. Install ffmpeg:
In the same Terminal window:
brew install ffmpeg
3. Open the release file
First, download the StoryToolkitAI zip file that fits with your platform, if you haven't done so already.
Unzip, copy the app file into your Applications folder, then double click and it should start.
4. Give the app all the necessary permissions to run
Mac OS might warn you of malicious software or it might ask you to allow the app to access files on your Mac. If the app won't start, first try to right click on the app and then click Open, and probably Open again on the dialog you'll see. If it still doesn't work, try this: Safely open apps on your Mac, and maybe this: control access to files and folders on Mac
If you get something like ""StoryToolkitAI.X.X.X" is damaged and can't be opened" and you downloaded the release from this page, open Terminal in the folder with the app and try to take the app out of Mac quarantine with this command before starting the app, for eg.:
xattr -d com.apple.quarantine ./StoryToolkitAI.0.17.5.M1.app. More details on the issue here.
Note: The standalone version does not open up the console. If you're interested to see what is happening behind the hood, open a terminal in the folder where you have the tool (typically in your Downloads or Applications folder) and run this command:
./[STORYTOOLKITAI_APP_NAME].app/Contents/MacOS/[STORYTOOLKIT_APP_NAME]
# (but replace [STORYTOOLKIT_APP_NAME] with the name of your release)
Windows
1. Download and install ffmpeg
The simplest approach is to use a package manager like Choco. Once you follow the installation steps for "Individual Use" on that page, you can install ffmpeg using this command in the command prompt:
choco install ffmpeg
2. Download and extract the release
The release file is a self extracting archive. Choose the folder where you want to have StoryToolkitAI on your computer and extract it there.
3. Open the app
Open the folder where you un-archived the file, and open the StoryToolkitAI.X.X.X.WIN.exe file, where X are the version numbers.
Info for all platforms
Make sure Resolve scripting is enabled
The tool only works with the Studio version of Davinci Resolve.
Davinci Resolve Preferences -> General, "External Scripting using", set to: Local
Downloading models to the cache
The first time you transcribe something with a specific model, it will take a bit longer to start the process because Whisper needs to download the model file (around 1.5GB for the medium model) on your local machine. But, after the model is saved on your machine, transcriptions will take less. The same applies to the advanced search function.
StoryToolkitAI v0.17.1 - super alpha
Download the standalone release here:
StoryToolkitAI 0.17.1 for Mac M1
StoryToolkitAI 0.17.1 for Mac Intel
StoryToolkitAI 0.17.1 for Windows (only for CUDA GPUs)
Tested only on a handful of machines, so please report if something doesn't work as expected.
What's new in this release
- Advanced Transcript Search Button allows you to search in multiple transcripts or even a whole folder of transcripts
- Switched to all-MiniLM-L6-v2 default model for significantly faster semantic searches
- Some new shortcuts, see them here
- FFMPEG check was buggy on some machines, so hopefully now it's fixed
Installation:
Mac OS
You need to install ffmpeg manually from the terminal and the easiest way to do that is via Homebrew:
1. Install Homebrew
Open up a Terminal window, then run:
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"
If Homebrew installation fails, first make sure you have xcode installed:
xcode-select --install
Note: if Homebrew installation fails for any reason, a good starting point is this page.
2. Install ffmpeg:
In the same Terminal window:
brew install ffmpeg
3. Open the release file
First, download the StoryToolkitAI zip file that fits with your platform, if you haven't done so already.
Unzip, copy the app file into your Applications folder, then double click and it should start.
4. Give the app all the necessary permissions to run
Mac OS might warn you of malicious software or it might ask you to allow the app to access files on your Mac. If the app won't start, first try to right click on the app and then click Open, and probably Open again on the dialog you'll see. If it still doesn't work, try this: Safely open apps on your Mac, and maybe this: control access to files and folders on Mac
If you get something like ""StoryToolkitAI.X.X.X" is damaged and can't be opened" and you downloaded the release from this page, open Terminal in the folder with the app and try to take the app out of Mac quarantine with this command before starting the app:
xattr -d com.apple.quarantine ./StoryToolkitAI.0.17.1.M1.app. More details on the issue here.
Note: The standalone version does not open up the console. If you're interested to see what is happening behind the hood, open a terminal in the folder where you have the tool (typically in your Downloads or Applications folder) and run this command:
./[STORYTOOLKITAI_APP_NAME].app/Contents/MacOS/[STORYTOOLKIT_APP_NAME]
# (but replace [STORYTOOLKIT_APP_NAME] with the name of your release)
Windows
1. Download and install ffmpeg
The simplest approach is to use a package manager like Choco. Once you follow the installation steps for "Individual Use" on that page, you can install ffmpeg using this command in the command prompt:
choco install ffmpeg
2. Download and extract the release
The release file is a self extracting archive. Choose the folder where you want to have StoryToolkitAI on your computer and extract it there.
3. Open the app
Open the folder where you un-archived the file, and open the StoryToolkitAI.X.X.X.WIN.exe file, where X are the version numbers.
Info for all platforms
Make sure Resolve scripting is enabled
The tool only works with the Studio version of Davinci Resolve.
Davinci Resolve Preferences -> General, "External Scripting using", set to: Local
Downloading models to the cache
The first time you transcribe something with a specific model, it will take a bit longer to start the process because Whisper needs to download the model file (around 1.5GB for the medium model) on your local machine. But, after the model is saved on your machine, transcriptions will take less. The same applies to the advanced search function._