octimot / storytoolkitai Goto Github PK

I love this app and it is indescribable how much time it saves me already.

I often find myself manually splitting the SRT bits in Resolve though, since I feel like the app often creates too much text on one item.
It would be great to set a max character limit in the interface!

I'm getting this following error when trying to open the M1 version (Have tested on M1 MacMini and M1 Max Macbook Pro)

Right clicking and selecting 'Open' doesn't work either.

hi. and using with resolve (with fantastic results in general), however, i have two interview subjects where story... chooses 'hebrew' as the lang. for some reason, (the interviewees are Israeli, but speaking english. is there a way to force story to always open in english?

thanks

well, just tried in stand-alone mode, and it's still choosing 'hebrew' as the default language. since i don't do any multilingual work, where can i set default to english?

many thanks and a very merry whatever ;-)

Is it possible to clear the queue?
I have tried to restart it, but every time it wants to continue a job that I actually want to abort.

Windows 10 Pro 21H2, python 3.9.13, Resolve Studio 18.1.1 w/ external scripting on Local. StoryToolkitAI installed via the INSTALLATION.md instructions, running in venv.

Running py StoryToolkitAI\app.py gives:

Traceback (most recent call last):
  File "J:\resolve\StoryToolkitAI\app.py", line 24, in <module>
    import librosa
ModuleNotFoundError: No module named 'librosa'

Tried pip install librosa as well; did not help.

I'm not a programmer, however, I have installed this successfully. When I run it, I receive the error:

SyntaxError: invalid syntax
Computer1:StoryToolkitAI-main computer1$ python app.py
Traceback (most recent call last):
File "app.py", line 14, in <module>
import mots_resolve
File "/Users/libraryfilmso4/Downloads/StoryToolkitAI-main/mots_resolve.py", line 176
marker_data = {**empty_marker_data, **marker_data} ^

I'd love to test this out, so any help would be awesome.

StoryToolkitAI How to use Large-v2 model？
I compared it StoryToolkitAI Large and whisper Large-v2,Two different outcomes,I Think StoryToolkitAI use old Large,How do I get StoryToolkitAI use Large-v2 model？

Would it be possible to have presets or a default setting for the Transcription Settings. Pretty much all of my transcriptions are going to be native English using CUDA etc. It would be nice to have that set as my default.

Hi there!

On the latest version I've found that the UI is having some strange things going on with it:

1. The transcription log is black and flickers:

2. The initial selection screen flickers, often the buttons turn totally white:

3. The dialogue that appears to select a model, language, etc. also was blacked out. Although the dropdowns were still visible (sorry no screenshot, currently running a translation).

StoryToolkit v0.16.14
macOS Monterey v12.4
iMac Pro 2017
3Ghz 10-Core Intel Xeon W
64 GB DDR4
Radeon Pro Vega 56 8GB VRAM

Dear all,

I installed ffmpeg using choco and tested it. I can run ffmpeg commands just fine. Unfortunately, when I try to run the StoryToolkitAI exe after extracting the folder with the binary for Windows 10, I only get the below screen flashing for a short while and disappearing.

Thanks so much for any advice. I need to admit I am not an expert user of Windows. Davinci Resolve 18.1 is installed the preferences are also set as advised, allowing local scripts to run.

Hi there,

I installed the app on a Windows 10 machine, got through all of the various steps, and when I run it I get this output:

Microsoft Windows [Version 10.0.19044.2130]
(c) Microsoft Corporation. All rights reserved.

C:\Users\jonat>cd "Start Menu"

C:\Users\jonat\Start Menu>cd ..

C:\Users\jonat>cd StoryToolkitAI

C:\Users\jonat\StoryToolkitAI>python3.9 app.py
Running StoryToolkit version 0.16.15
Using cuda for Torch / Whisper.

C:\Users\jonat\StoryToolkitAI>

It seems to run, then there's a long delay, and then the prompt returns with no activity indicating that it's working. I have Resolve open and added to my environment variables.

Thanks!

Hello dears,

I have been using StoryToolkitAI inside my DaVinci Resolve Studio 18 on a M1/16GB RAM MacBookPro and loving the transcription-translation results from Portuguese into English. Thank you for putting so much work into this wonderful API.

Despite the high match results for generic language, content with unique words, such as people and place names, disciplinary scientific jargon, the machine-transcription and translation of course needs a lot of human correction.

My question to the developers and community is about efficient ways to correct the original language texts and EN files in a way that you can somehow automate or semi-automate the correction process. In other words, I am trying to avoid the painstaking process of correcting words in all 4 files separately... the two files: srt and txt in original language and the English language srt and txt.

Also, after reading all instructions I am still a bit confused about the difference among running "Transcribe Timeline" first, then "Translate timeline to English"... in addition to having the option to select a "transcribe+translate" in the window that pops up upon selecting "Translate timeline to English". If I do "transcribe+translate" for a 10-min, video, I am getting both PT and EN files in about 17-min. which I think is really good, right?

Thank you so much for your replies!

Wishing everyone a healthy and happy 2023!
Vanessa

I have installed everything, and the cmd screen says i'm Running StoryToolkitAI version 0.17.6.2. Using cpu for Torch /Whisper.
But i don't get any fancy 2000 era gui opening.

I'm not sure what to do next.

Running windows 10 with Nvidia RTX A5000

This might be related to another issue here but I seem to be getting enough differences that I'm doing this as a new issue.

I am very new to all of this Python and coding stuff so I might have messed something up with the instillation, but it was all followed correctly to my knowledge. This is all on windows 10. I was also running all of the commands in command prompt because when I tried in Python it kept giving me syntax errors. I also made my virtual environment with a different command than what is shown in the installation guide. py -m venv (environment name) was the command I used in command prompt. When I tried what was suggested, both in and out of python, this is what I got:

C:\video editing\Storytoolkit>python3.9 virtualenv -p python3.9 venv 'python3.9' is not recognized as an internal or external command, operable program or batch file.

C:\video editing\Storytoolkit>py Python 3.9.13 (tags/v3.9.13:6de2ca5, May 17 2022, 16:36:42) [MSC v.1929 64 bit (AMD64)] on win32 Type "help", "copyright", "credits" or "license" for more information.
>>> python3.9 virtualenv -p python3.9 venv
File "<stdin>", line 1
python3.9 virtualenv -p python3.9 venv
^
SyntaxError: invalid syntax

Not sure if any of the above is actually responsible for the this issue or not but I included it to be safe

For the issue itself, whenever I run the command python StoryToolkitAI/app.py in my virtual environment, I get this:

(transcript) C:\video editing\Storytoolkit>python StoryToolkitAI/app.py
INFO: Running StoryToolkitAI version 0.16.16
INFO: Using cpu for Torch / Whisper.
Unable to find module DaVinciResolveScript from $PYTHONPATH - trying default locations
Unable to find module DaVinciResolveScript - please ensure that the module DaVinciResolveScript is discoverable by python
For a default DaVinci Resolve installation, the module is expected to be located in: C:\ProgramData\Blackmagic Design\DaVinci Resolve\Support\Developer\Scripting\Modules\
Traceback (most recent call last):
File "C:\video editing\Storytoolkit\StoryToolkitAI\python_get_resolve.py", line 13, in GetResolve import DaVinciResolveScript as bmd
ModuleNotFoundError: No module named 'DaVinciResolveScript'

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "C:\video editing\Storytoolkit\StoryToolkitAI\python_get_resolve.py", line 27, in GetResolve bmd = imp.load_source('DaVinciResolveScript', expectedPath+"DaVinciResolveScript.py")
File "C:\Users\username\AppData\Local\Programs\Python\Python39\lib\imp.py", line 171, in load_source module = _load(spec)
File "<frozen importlib._bootstrap>", line 711, in _load
File "<frozen importlib._bootstrap>", line 680, in _load_unlocked
File "<frozen importlib._bootstrap_external>", line 850, in exec_module
File "<frozen importlib._bootstrap>", line 228, in _call_with_frames_removed
File "C:\ProgramData\Blackmagic Design\DaVinci Resolve\Support\Developer\Scripting\Modules\DaVinciResolveScript.py", line 36, in <module>
raise ImportError("Could not locate module dependencies")
ImportError: Could not locate module dependencies

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "C:\video editing\Storytoolkit\StoryToolkitAI\app.py", line 3877, in poll_resolve_data resolve_data = mots_resolve.get_resolve_data(silent=True)
File "C:\video editing\Storytoolkit\StoryToolkitAI\mots_resolve.py", line 89, in get_resolve_data = initialize_resolve(silent=silent)
File "C:\video editing\Storytoolkit\StoryToolkitAI\mots_resolve.py", line 34, in initialize_resolve resolve = GetResolve()
File "C:\video editing\Storytoolkit\StoryToolkitAI\python_get_resolve.py", line 32, in GetResolve sys.exit()
SystemExit

WARNING: Resolve is not reachable.
INFO: Starting StoryToolkitAI GUI

Sorry if this is a lot of info, but I'm not sure what is important and what isn't.

I like the idea of being able to use the already-rendered, audio-only file from a first transcription pass to do a new transcription. The problem seems to be that the auto-render that StoryToolkitAI does generates an (audio-only) .mov file, and the dialog that comes up when you Shift-click the Transcribe Timeline button is filtered to only .mp4, .wav, .mp3, so the existing .mov file cannot be selected.

Didn't expect enter key to submit this, I wasn't done writing. Just did a fresh install of python 3.9.12. Running Windows 11 and Resolve 18.1.1. The UI for transcription will open, however the app can't connect to the timeline. CUDA edition of the app.

Tried running 0.17.5 and it would not connect to Resolve because it's not where Toolkit thinks it should be, and I haven't found anything to let me change the path it's looking at. Here's a screenshot of the exact issue:

I still managed to transcribe it by feeding Toolkit the audio file. This issue is absent from 0.17.1

As an aside, I have noticed that the transcripts across both versions have been identical, but that they has been a variety of small mistakes that I don't think it should be getting. It's more noticeable and understandable when I speak, as I tend to stutter and trip over my words a bit, but it happens to the others in the recording quite a bit as well. I'm not sure if it has something to do with audio quality or the length of the transcription or what, but it's something I feel that it's worth mentioning. I can other up the json file and video if you want so you can see for yourself. For context it's a dnd session that I'm transcribing, so it makes sense for some of the words to be wrong but not most of the ones that are.

With the transcript window open and "Keep on Top" enabled, any button that would pop out a choose box requiring further actions -- e.g. pressing 'T' to re-transcribe -- the choose box would appear behind the main window and there is no indication to the user this is happening, other than the OS error alert sound when the user tries to click somewhere on the screen (it would appear as if the app is unresponsive, but it's not -- it's just stuck with the choose box appearing behind the window)

Please read this first

FFMPEG is a software that we use to convert audio before feeding in to the transcription models. We have inherited this process from the OpenAI Whisper project, since this is what they use to convert audio.

However, for most use cases in StoryToolkitAI, FFMPEG is not needed.

In other words, you only need it on your machine to benefit from the absolute full functionality of StoryToolkitAI (meaning the last 1%) and only if you're planning to transcribe files that don't use the Linear PCM codec for their audio. As a matter of fact, other formats work without ffmpeg aswell, but we just haven't tested it widely enough to know for sure which ones.

What this means for you

If you're like most users and don't have to deal with very specific file codecs and formats, 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. Also, make sure that you're audio is mono, to prevent the algorithm from ignoring audio from multiple channels. A if this is good enough for you just ignore the FFMPEG was not found on this machine message.

Let us know what you think

We're planning to remove the error at startup in a future release and have it pop up when FFMPEG is actually needed.

If you think this is a mistake or dealing with non-WAV Linear PCM files is more common than we think, just let us know here.

i try to guide install homebrew and 1~7
i struggle to this step 7

pip install git+https://github.com/openai/whisper.git
============================================
Cloning https://github.com/openai/whisper.git to /private/var/folders/cy/cphr3x4j2cs0tj5wt3rt72jm0000gn/T/pip-req-build-uf_5g3ke
Running command git clone --filter=blob:none --quiet https://github.com/openai/whisper.git /private/var/folders/cy/cphr3x4j2cs0tj5wt3rt72jm0000gn/T/pip-req-build-uf_5g3ke
Resolved https://github.com/openai/whisper.git to commit 28769fcfe50755a817ab922a7bc83483159600a9
Preparing metadata (setup.py) ... done
Collecting numpy
Using cached numpy-1.24.1-cp39-cp39-macosx_11_0_arm64.whl (13.9 MB)
Collecting torch
Using cached torch-1.13.1-cp39-none-macosx_11_0_arm64.whl (53.2 MB)
Collecting tqdm
Downloading tqdm-4.64.1-py2.py3-none-any.whl (78 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 78.5/78.5 kB 678.7 kB/s eta 0:00:00
Collecting more-itertools
Downloading more_itertools-9.0.0-py3-none-any.whl (52 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 52.8/52.8 kB 774.9 kB/s eta 0:00:00
Collecting transformers>=4.19.0
Downloading transformers-4.25.1-py3-none-any.whl (5.8 MB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.8/5.8 MB 1.6 MB/s eta 0:00:00
Collecting ffmpeg-python==0.2.0
Downloading ffmpeg_python-0.2.0-py3-none-any.whl (25 kB)
Collecting future
Downloading future-0.18.2.tar.gz (829 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 829.2/829.2 kB 2.3 MB/s eta 0:00:00
Preparing metadata (setup.py) ... done
Collecting huggingface-hub<1.0,>=0.10.0
Downloading huggingface_hub-0.11.1-py3-none-any.whl (182 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 182.4/182.4 kB 1.7 MB/s eta 0:00:00
Collecting regex!=2019.12.17
Downloading regex-2022.10.31-cp39-cp39-macosx_11_0_arm64.whl (287 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 287.2/287.2 kB 1.9 MB/s eta 0:00:00
Collecting tokenizers!=0.11.3,<0.14,>=0.11.1
Downloading tokenizers-0.13.2.tar.gz (359 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 359.1/359.1 kB 2.4 MB/s eta 0:00:00
Installing build dependencies ... done
Getting requirements to build wheel ... done
Preparing metadata (pyproject.toml) ... done
Collecting filelock
Downloading filelock-3.9.0-py3-none-any.whl (9.7 kB)
Collecting requests
Using cached requests-2.28.1-py3-none-any.whl (62 kB)
Collecting pyyaml>=5.1
Downloading PyYAML-6.0-cp39-cp39-macosx_11_0_arm64.whl (173 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 174.0/174.0 kB 3.1 MB/s eta 0:00:00
Collecting packaging>=20.0
Downloading packaging-22.0-py3-none-any.whl (42 kB)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 42.6/42.6 kB 1.1 MB/s eta 0:00:00
Collecting typing-extensions
Using cached typing_extensions-4.4.0-py3-none-any.whl (26 kB)
Collecting urllib3<1.27,>=1.21.1
Using cached urllib3-1.26.13-py2.py3-none-any.whl (140 kB)
Collecting certifi>=2017.4.17
Using cached certifi-2022.12.7-py3-none-any.whl (155 kB)
Collecting idna<4,>=2.5
Using cached idna-3.4-py3-none-any.whl (61 kB)
Collecting charset-normalizer<3,>=2
Using cached charset_normalizer-2.1.1-py3-none-any.whl (39 kB)
Building wheels for collected packages: whisper, tokenizers, future
Building wheel for whisper (setup.py) ... done
Created wheel for whisper: filename=whisper-1.0-py3-none-any.whl size=1175322 sha256=aede5db8b3053ea1e5472ae6348cfcc36195fbce102060ab1fd1761680aec242
Stored in directory: /private/var/folders/cy/cphr3x4j2cs0tj5wt3rt72jm0000gn/T/pip-ephem-wheel-cache-c77i_xox/wheels/fe/03/29/e7919208d11b4ab32972cb448bb84a9a675d92cd52c9a48341
Building wheel for tokenizers (pyproject.toml) ... error
error: subprocess-exited-with-error

× Building wheel for tokenizers (pyproject.toml) did not run successfully.
│ exit code: 1
╰─> [51 lines of output]
running bdist_wheel
running build
running build_py
creating build
creating build/lib.macosx-11.0-arm64-cpython-39
creating build/lib.macosx-11.0-arm64-cpython-39/tokenizers
copying py_src/tokenizers/init.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers
creating build/lib.macosx-11.0-arm64-cpython-39/tokenizers/models
copying py_src/tokenizers/models/init.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/models
creating build/lib.macosx-11.0-arm64-cpython-39/tokenizers/decoders
copying py_src/tokenizers/decoders/init.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/decoders
creating build/lib.macosx-11.0-arm64-cpython-39/tokenizers/normalizers
copying py_src/tokenizers/normalizers/init.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/normalizers
creating build/lib.macosx-11.0-arm64-cpython-39/tokenizers/pre_tokenizers
copying py_src/tokenizers/pre_tokenizers/init.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/pre_tokenizers
creating build/lib.macosx-11.0-arm64-cpython-39/tokenizers/processors
copying py_src/tokenizers/processors/init.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/processors
creating build/lib.macosx-11.0-arm64-cpython-39/tokenizers/trainers
copying py_src/tokenizers/trainers/init.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/trainers
creating build/lib.macosx-11.0-arm64-cpython-39/tokenizers/implementations
copying py_src/tokenizers/implementations/byte_level_bpe.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/implementations
copying py_src/tokenizers/implementations/sentencepiece_unigram.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/implementations
copying py_src/tokenizers/implementations/sentencepiece_bpe.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/implementations
copying py_src/tokenizers/implementations/base_tokenizer.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/implementations
copying py_src/tokenizers/implementations/init.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/implementations
copying py_src/tokenizers/implementations/char_level_bpe.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/implementations
copying py_src/tokenizers/implementations/bert_wordpiece.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/implementations
creating build/lib.macosx-11.0-arm64-cpython-39/tokenizers/tools
copying py_src/tokenizers/tools/init.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/tools
copying py_src/tokenizers/tools/visualizer.py -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/tools
copying py_src/tokenizers/init.pyi -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers
copying py_src/tokenizers/models/init.pyi -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/models
copying py_src/tokenizers/decoders/init.pyi -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/decoders
copying py_src/tokenizers/normalizers/init.pyi -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/normalizers
copying py_src/tokenizers/pre_tokenizers/init.pyi -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/pre_tokenizers
copying py_src/tokenizers/processors/init.pyi -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/processors
copying py_src/tokenizers/trainers/init.pyi -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/trainers
copying py_src/tokenizers/tools/visualizer-styles.css -> build/lib.macosx-11.0-arm64-cpython-39/tokenizers/tools
running build_ext
running build_rust
error: can't find Rust compiler

  If you are using an outdated pip version, it is possible a prebuilt wheel is available for this package but pip is not able to install from it. Installing from the wheel would avoid the need for a Rust compiler.
  
  To update pip, run:
  
      pip install --upgrade pip
  
  and then retry package installation.
  
  If you did intend to build this package from source, try installing a Rust compiler from your system package manager and ensure it is on the PATH during installation. Alternatively, rustup (available at https://rustup.rs) is the recommended way to download and update the Rust compiler toolchain.
  [end of output]

note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for tokenizers
Building wheel for future (setup.py) ... done
Created wheel for future: filename=future-0.18.2-py3-none-any.whl size=491059 sha256=3887bdef91b6279f6d5c9b3d1f7450d429f16c6b49dd27814cda2e555f562999
Stored in directory: /Users/hanjungu/Library/Caches/pip/wheels/96/66/19/2de75120f5d0bc185e9d16cf0fd223d8471ed025de08e45867
Successfully built whisper future
Failed to build tokenizers
ERROR: Could not build wheels for tokenizers, which is required to install pyproject.toml-based projects

All the previous steps worked alright.

Encountering this issue after installing everything and running python StoryToolkitAI/app.py

Traceback (most recent call last): File "/Users/officeeditor1/Movies/StoryToolKitAI/StoryToolkitAI/app.py", line 14, in <module> import mots_resolve File "/Users/officeeditor1/Movies/StoryToolKitAI/StoryToolkitAI/mots_resolve.py", line 15, in <module> from timecode import Timecode ModuleNotFoundError: No module named 'timecode'

Any suggestions?

"Copy Timeline Markers to Same Clip" copies markers from the open timeline in the record monitor to the timeline clip itself.

However, it would be great to have a functions which copies the markers to the source clips in the media pool which make up the timeline ("Copy markers from timeline to timeline source clips").

This would allow to keep the transcript/translation on the source material, so that the source material may be re-edited into new timelines, without loosing the marker info.

Hey guys,

The app won't open up for me. I launched storytoolkitai (standalone) but the command line keeps disappearing after these messages in the pic. No difference if i launch it with admin privilege

This is what I can see in the console:

2022-12-01 20:45:27.200 StoryToolkitAI.0.17.5.M1[73259:2499663] +[CATransaction synchronize] called within transaction
INFO: Saving into /Users/antonioparraga/Desktop and starting render.
Exception in Tkinter callback
Traceback (most recent call last):
  File "/Users/antonioparraga/Desktop/StoryToolkitAI.0.17.5.M1.app/Contents/Resources/lib/python3.9/tkinter/__init__.py", line 1892, in __call__
  File "/Users/antonioparraga/Desktop/StoryToolkitAI.0.17.5.M1.app/Contents/Resources/app.py", line 2794, in <lambda>
  File "/Users/antonioparraga/Desktop/StoryToolkitAI.0.17.5.M1.app/Contents/Resources/app.py", line 6084, in prepare_transcription_file
  File "mots_resolve.pyc", line 985, in render_timeline
TypeError: 'NoneType' object is not callable

Any help could be appreciate

Any suggestions here, as when hitting "Translate Audio"
in StoryToolkitAS v0.17.6.2 with Resolve 18, and Mac Monterey 12.5.1 ?
I get "A Resolve Timeline is not available ..."

I've tried to transcribe a timeline a few times now. Every time it goes to do so, it gives me this error back:

INFO: Starting to transcribe Timeline 1
INFO: Using cuda for Torch / Whisper.
INFO: Transcribing Timeline 1.
This may take a while.
librosa\util\decorators.py:88: UserWarning: PySoundFile failed. Trying audioread instead. return f(*args, **kwargs)
ERROR: Traceback (most recent call last):
File "app.py", line 4659, in transcribe_from_queue
File "app.py", line 5011, in whisper_transcribe
File "app.py", line 4755, in whisper_transcribe_segments
File "whisper\transcribe.py", line 181, in transcribe result: DecodingResult = decode_with_fallback(segment)
File "whisper\transcribe.py", line 117, in decode_with_fallback decode_result = model.decode(segment, options)
File "torch\autograd\grad_mode.py", line 27, in decorate_context return func(*args, **kwargs)
File "whisper\decoding.py", line 701, in decode result = DecodingTask(model, options).run(mel)
File "torch\autograd\grad_mode.py", line 27, in decorate_context return func(*args, **kwargs)
File "whisper\decoding.py", line 617, in run audio_features: Tensor = self._get_audio_features(mel) # encoder forward pass
File "whisper\decoding.py", line 561, in _get_audio_features audio_features = self.model.encoder(mel)
File "torch\nn\modules\module.py", line 1130, in _call_impl return forward_call(*input, **kwargs)
File "whisper\model.py", line 156, in forward x = block(x)
File "torch\nn\modules\module.py", line 1130, in _call_impl return forward_call(*input, **kwargs)
File "whisper\model.py", line 124, in forward x = x + self.attn(self.attn_ln(x), mask=mask, kv_cache=kv_cache)
File "torch\nn\modules\module.py", line 1130, in _call_impl return forward_call(*input, **kwargs)
File "whisper\model.py", line 73, in forward q = self.query(x)
File "torch\nn\modules\module.py", line 1130, in _call_impl return forward_call(*input, **kwargs)
File "whisper\model.py", line 36, in forward return F.linear(
RuntimeError: CUDA error: CUBLAS_STATUS_ALLOC_FAILED when calling ``cublasCreate(handle)``

For the last line, it's just one ` on either side of cublas. Formatting was a bit weird there.

From what I can tell, this is all on whispers side correct? It's failing to allocate the cuda correctly, even though I am using only the medium model? Would it be advised for me to try it with only Davinci and storykit open to eliminate other applications drawing resources away as a factor?

Hi, I've just installed StoryToolkitAI on Windows 10 by following your detailed instruction. Inside the virtual environment, I ran the app. Then it shows this:

INFO: Running StoryToolkitAI version 0.16.16
INFO: Using cuda for Torch / Whisper.
Unable to find module DaVinciResolveScript from $PYTHONPATH - trying default locations

I'm quite new to running app with Python (I'm using Python 3.9 from Windows Store), did I miss something? My Davinci Resolve Studio's external scripting setting is set to Local

Usually everything worked as expected, today despite all dependencies being met the programme stopped starting.

Requirement already satisfied: certifi==2022.9.24 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 1)) (2022.9.24)
Requirement already satisfied: charset-normalizer==2.1.1 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 2)) (2.1.1)
Requirement already satisfied: ffmpeg-python==0.2.0 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 3)) (0.2.0)
Requirement already satisfied: filelock==3.8.0 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 4)) (3.8.0)
Requirement already satisfied: future==0.18.2 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 5)) (0.18.2)
Requirement already satisfied: huggingface-hub>=0.10.0 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 6)) (0.10.0)
Requirement already satisfied: idna==3.4 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 7)) (3.4)
Requirement already satisfied: librosa>=0.9.1 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 8)) (0.9.2)
Requirement already satisfied: more-itertools>=8.14.0 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 9)) (8.14.0)
Requirement already satisfied: numpy in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 10)) (1.23.1)
Requirement already satisfied: packaging==21.3 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 11)) (21.3)
Requirement already satisfied: pyparsing==3.0.9 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 12)) (3.0.9)
Requirement already satisfied: PyYAML==6.0 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 13)) (6.0)
Requirement already satisfied: regex==2022.9.13 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 14)) (2022.9.13)
Requirement already satisfied: requests==2.28.1 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 15)) (2.28.1)
Requirement already satisfied: setuptools in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 16)) (65.4.1)
Requirement already satisfied: sentence_transformers in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 17)) (2.2.2)
Requirement already satisfied: timecode==1.3.1 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 18)) (1.3.1)
Requirement already satisfied: tokenizers==0.12.1 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 19)) (0.12.1)
Requirement already satisfied: torch in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 20)) (1.12.1+cu116)
Requirement already satisfied: tqdm in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 21)) (4.64.1)
Requirement already satisfied: transformers>=4.19.0 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 22)) (4.22.2)
Requirement already satisfied: typing_extensions==4.3.0 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 23)) (4.3.0)
Requirement already satisfied: urllib3==1.26.12 in c:\users\bagro.platformio\penv\lib\site-packages (from -r .\requirements.txt (line 24)) (1.26.12)
Requirement already satisfied: soundfile>=0.10.2 in c:\users\bagro.platformio\penv\lib\site-packages (from librosa>=0.9.1->-r .\requirements.txt (line 8)) (0.11.0)
Requirement already satisfied: decorator>=4.0.10 in c:\users\bagro.platformio\penv\lib\site-packages (from librosa>=0.9.1->-r .\requirements.txt (line 8)) (5.1.1)
Requirement already satisfied: numba>=0.45.1 in c:\users\bagro.platformio\penv\lib\site-packages (from librosa>=0.9.1->-r .\requirements.txt (line 8)) (0.56.4)
Requirement already satisfied: scipy>=1.2.0 in c:\users\bagro.platformio\penv\lib\site-packages (from librosa>=0.9.1->-r .\requirements.txt (line 8)) (1.9.3)
Requirement already satisfied: resampy>=0.2.2 in c:\users\bagro.platformio\penv\lib\site-packages (from librosa>=0.9.1->-r .\requirements.txt (line 8)) (0.4.2)
Requirement already satisfied: audioread>=2.1.9 in c:\users\bagro.platformio\penv\lib\site-packages (from librosa>=0.9.1->-r .\requirements.txt (line 8)) (3.0.0)
Requirement already satisfied: scikit-learn>=0.19.1 in c:\users\bagro.platformio\penv\lib\site-packages (from librosa>=0.9.1->-r .\requirements.txt (line 8)) (1.1.3)
Requirement already satisfied: pooch>=1.0 in c:\users\bagro.platformio\penv\lib\site-packages (from librosa>=0.9.1->-r .\requirements.txt (line 8)) (1.6.0)
Requirement already satisfied: joblib>=0.14 in c:\users\bagro.platformio\penv\lib\site-packages (from librosa>=0.9.1->-r .\requirements.txt (line 8)) (1.2.0)
Requirement already satisfied: sentencepiece in c:\users\bagro.platformio\penv\lib\site-packages (from sentence_transformers->-r .\requirements.txt (line 17)) (0.1.97)
Requirement already satisfied: nltk in c:\users\bagro.platformio\penv\lib\site-packages (from sentence_transformers->-r .\requirements.txt (line 17)) (3.7)
Requirement already satisfied: torchvision in c:\users\bagro.platformio\penv\lib\site-packages (from sentence_transformers->-r .\requirements.txt (line 17)) (0.13.1+cu116)
Requirement already satisfied: colorama in c:\users\bagro.platformio\penv\lib\site-packages (from tqdm->-r .\requirements.txt (line 21)) (0.4.5)
Requirement already satisfied: llvmlite<0.40,>=0.39.0dev0 in c:\users\bagro.platformio\penv\lib\site-packages (from numba>=0.45.1->librosa>=0.9.1->-r .\requirements.txt (line 8)) (0.39.1)
Requirement already satisfied: appdirs>=1.3.0 in c:\users\bagro.platformio\penv\lib\site-packages (from pooch>=1.0->librosa>=0.9.1->-r .\requirements.txt (line 8)) (1.4.4)
Requirement already satisfied: threadpoolctl>=2.0.0 in c:\users\bagro.platformio\penv\lib\site-packages (from scikit-learn>=0.19.1->librosa>=0.9.1->-r .\requirements.txt (line 8)) (3.1.0)
Requirement already satisfied: cffi>=1.0 in c:\users\bagro.platformio\penv\lib\site-packages (from soundfile>=0.10.2->librosa>=0.9.1->-r .\requirements.txt (line 8)) (1.15.1)
Requirement already satisfied: click in c:\users\bagro.platformio\penv\lib\site-packages (from nltk->sentence_transformers->-r .\requirements.txt (line 17)) (8.1.3)
Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in c:\users\bagro.platformio\penv\lib\site-packages (from torchvision->sentence_transformers->-r .\requirements.txt (line 17)) (9.2.0)
Requirement already satisfied: pycparser in c:\users\bagro.platformio\penv\lib\site-packages (from cffi>=1.0->soundfile>=0.10.2->librosa>=0.9.1->-r .\requirements.txt (line 8)) (2.21)

C:\Program Files\Blackmagic Design_StoryToolkitAI> py app.py
Traceback (most recent call last):
File "C:\Program Files\Blackmagic Design_StoryToolkitAI\app.py", line 20, in
import torch
ModuleNotFoundError: No module named 'torch'

C:\Program Files\Blackmagic Design_StoryToolkitAI> pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu116
Looking in indexes: https://pypi.org/simple, https://download.pytorch.org/whl/cu116
Requirement already satisfied: torch in c:\users\bagro.platformio\penv\lib\site-packages (1.12.1+cu116)
Requirement already satisfied: torchvision in c:\users\bagro.platformio\penv\lib\site-packages (0.13.1+cu116)
Requirement already satisfied: torchaudio in c:\users\bagro.platformio\penv\lib\site-packages (0.12.1+cu116)
Requirement already satisfied: typing-extensions in c:\users\bagro.platformio\penv\lib\site-packages (from torch) (4.3.0)
Requirement already satisfied: requests in c:\users\bagro.platformio\penv\lib\site-packages (from torchvision) (2.28.1)
Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in c:\users\bagro.platformio\penv\lib\site-packages (from torchvision) (9.2.0)
Requirement already satisfied: numpy in c:\users\bagro.platformio\penv\lib\site-packages (from torchvision) (1.23.1)
Requirement already satisfied: certifi>=2017.4.17 in c:\users\bagro.platformio\penv\lib\site-packages (from requests->torchvision) (2022.9.24)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in c:\users\bagro.platformio\penv\lib\site-packages (from requests->torchvision) (1.26.12)
Requirement already satisfied: idna<4,>=2.5 in c:\users\bagro.platformio\penv\lib\site-packages (from requests->torchvision) (3.4)
Requirement already satisfied: charset-normalizer<3,>=2 in c:\users\bagro.platformio\penv\lib\site-packages (from requests->torchvision) (2.1.1)

I ran a 1 hour INTV through this using the "large" option. I set my language to Portuguese, and when it finished it presented me with a Portuguese transcript, rather than an English one.

Here's my Terminal information:

(venv) LibraryFilmsO26:~ LibraryFilmsO26$ python StoryToolkitAI/app.py 
Running StoryToolkit version 0.16.14
Using cpu for Torch / Whisper.
Unable to find module DaVinciResolveScript from $PYTHONPATH - trying default locations
Starting StoryToolkitAI GUI
Prompting user for render path.
Project settings directory /Users/LibraryFilmsO26/StoryToolkitAI/projects/ai test doesn't exist.
Creating project settings directory.
Updated project settings file /Users/LibraryFilmsO26/StoryToolkitAI/projects/ai test/project.json.
Saving into /Volumes/LF_SSD_03/AI TEST/KOXA and starting render.
Starting to transcribe GMAC3_20211130_NAZARE_Rodrigo Koxa_INTV
Using cpu for Torch / Whisper.
Currently transcribing: GMAC3_20211130_NAZARE_Rodrigo Koxa_INTV
Loading Whisper large model.
Selected Whisper model is multilingual.
Transcribing GMAC3_20211130_NAZARE_Rodrigo Koxa_INTV.
This may take a while.
/Users/LibraryFilmsO26/venv/lib/python3.9/site-packages/whisper/transcribe.py:78: UserWarning: FP16 is not supported on CPU; using FP32 instead
  warnings.warn("FP16 is not supported on CPU; using FP32 instead")
Finished transcription for GMAC3_20211130_NAZARE_Rodrigo Koxa_INTV in 43952 seconds
Config setting transcripts_always_on_top saved as False 
Updated config file /Users/LibraryFilmsO26/StoryToolkitAI/config.json with transcripts_always_on_top data.

Thanks!

Hey,

first of all awesome work, and huge thanks for making this Open Source. I got two questions:

1. Is it possible to export/import the transcript as text+ objects? (creating via the api)
   Why? Text+ has way more features to stylize text (only for burn in ofc)
   How? My idea would be, to create a new video track (to avoid overwriting existing stuff), and create all text+ objects there

2. I've skimmed over the code, but couldn't really find where to set/rewrite a bit code to set the length of each "line"/segment. Will there be an option in the future for this or a similar function?

I'll try to dig a bit deeper and maybe find a solution, but maybe its already in the works, which would be awesome

Managed to get this working today and it is utterly brilliant! Thankyou.
Would it be possible to add Timeline or Source TC to the Transcribed file so that when I pass it to an Edit Producer or Director etc. They can pull sections on a script with the TC. Then I can go to the edit and easily find sections by TC.

This is how I have done this before on Trint etc. and works pretty well. Unless you have a better way!

thanks again
Nick

So just read this really cool word-level timed transcription openai/whisper#684 and I though maybe that could be implemented so that when editing, you can click a word in the Transcription Log then it jumps to that specific word.

Mainly thinking about this workflow so that it's more similar to Descript where you edit the timeline through phrases and words instead of scrubbing through the actual timeline. Right now, the issue is of course when you make an edit in the resolve timeline, that section doesn't get updated in the log. Then the log gets desynced because the timings are now different compared to the timeline. So I thought maybe we could delete the words in the log then sync the deleted words' timecodes to the resolve timeline.

Currently, I just set it up where I only use the log for navigation and when I cut down stuff from the timeline, I retain the gaps so the log stays in sync.

The advanced search is really cool for this one too: you can type in "repeating phrases" then it would spit out the times the speakers repeat and rephrase themselves, so you can clearly see which parts need some editing. Unfortunate that it only shows up the top 10 results, so maybe another feature to have a "Load more" button or maybe it reactively loads more results once you scroll down more.

v 0.17.5.WIN, from expanded standalone. Feels like there ought to be more to the setup than just expanding the file...
So - what Transcribe button!? (Screencap attached.) All I get is a blank Powershell/Terminal window. Have uninstalled all instances of Python, and tried clean installs of both 3.11.1 and 3.9,12, all to no avail.
Have set external scripting use to local and am on Studio v 18.1.2
Windows 11, all up to date.
Only unusual thing is that I'm running a dual monitor setup.
Any thoughts?
Was very excited to try this...

I get this window flashing up on the screen when trying to open the .exe file. Any ideas? Am I missing some kind of Python library?

Getting CUDA out of memory errors as follows:

ERROR: Traceback (most recent call last):
File "app.py", line 4659, in transcribe_from_queue
File "app.py", line 4918, in whisper_transcribe
File "whisper_init_.py", line 113, in load_model
return model.to(device)
File "torch\nn\modules\module.py", line 927, in to
return self._apply(convert)
File "torch\nn\modules\module.py", line 579, in _apply
module._apply(fn)
File "torch\nn\modules\module.py", line 579, in _apply
module._apply(fn)
File "torch\nn\modules\module.py", line 579, in _apply
module._apply(fn)
[Previous line repeated 2 more times]
File "torch\nn\modules\module.py", line 602, in _apply
param_applied = fn(param)
File "torch\nn\modules\module.py", line 925, in convert
return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None, non_blocking)
RuntimeError: CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 4.00 GiB total capacity; 3.46 GiB already allocated; 0 bytes free; 3.47 GiB reserved in total by PyTorch) If reserved memory is >> allocated memory try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF

SYSTEM DETAILS ATTACHED

DxDiag.txt

When transcribing got audio rendering problem. It renders a white noise clip !
But when i re-render it in manual, it's ok and then i can transcribe at the second step of the script procedure !

on a DVR 18.1.1
Mac Intel version

Trying to install StoryToolkitAI on a new computer and run into setup issues.

Turned out that OpenAI renamed the package to openai-whisper.

Log file in the

c:\Users[USER NAME]\StoryToolkitAI\app.log

is generated with some error: After block of information before adding next block is added
2022-11-15 11:43:02,744 - StAI - DEBUG:
without any data.

2022-11-15 11:42:20,770 - StAI - INFO: FFMPEG found at C:\ProgramData\chocolatey\bin\ffmpeg.EXE (app.py:6922)
2022-11-15 11:42:20,785 - StAI - INFO: Using cuda for Torch / Whisper. (app.py:4446)
2022-11-15 11:42:20,785 - StAI - DEBUG: MotsResolve module initialized. (mots_resolve.py:42)
2022-11-15 11:42:20,801 - StAI - INFO: Resolve API module found and loaded (mots_resolve.py:177)
2022-11-15 11:43:02,744 - StAI - DEBUG:

Platform: Windows 10
Platform version: 10.0.22621
OS: 10 10.0.22621 SP0 Multiprocessor Free ('', '', '')
running Python 3.9.13.final.0
-------------- (app.py:140)
2022-11-15 11:43:03,120 - StAI - DEBUG: All package requirements met. (app.py:160)
2022-11-15 11:43:03,122 - StAI - INFO: �[1m�[4mRunning StoryToolkitAI version 0.17.4 (app.py:6464)
2022-11-15 11:43:03,218 - StAI - DEBUG: Looking for ffmpeg in env variable. (app.py:6895)
2022-11-15 11:43:03,218 - StAI - DEBUG: FFMPEG_BINARY env variable is empty. Looking for ffmpeg in PATH. (app.py:6902)
2022-11-15 11:43:03,220 - StAI - DEBUG: Checking ffmpeg binary: C:\ProgramData\chocolatey\bin\ffmpeg.EXE (app.py:6914)
2022-11-15 11:43:03,256 - StAI - DEBUG: FFMPEG exit code: 1 (app.py:6919)
2022-11-15 11:43:03,256 - StAI - INFO: FFMPEG found at C:\ProgramData\chocolatey\bin\ffmpeg.EXE (app.py:6922)
2022-11-15 11:43:08,098 - StAI - INFO: Using cuda for Torch / Whisper. (app.py:4446)
2022-11-15 11:43:08,099 - StAI - DEBUG: MotsResolve module initialized. (mots_resolve.py:42)
2022-11-15 11:43:08,114 - StAI - INFO: Resolve API module found and loaded (mots_resolve.py:177)
2022-11-15 11:48:13,738 - StAI - DEBUG:

Platform: Windows 10
Platform version: 10.0.22621
OS: 10 10.0.22621 SP0 Multiprocessor Free ('', '', '')
running Python 3.9.13.final.0
-------------- (app.py:140)
2022-11-15 11:48:14,124 - StAI - DEBUG: All package requirements met. (app.py:160)
2022-11-15 11:48:14,124 - StAI - INFO: �[1m�[4mRunning StoryToolkitAI version 0.17.4 (app.py:6464)
2022-11-15 11:48:14,398 - StAI - DEBUG: Looking for ffmpeg in env variable. (app.py:6895

Like:

2022-11-15 11:43:08,114 - StAI - INFO: Resolve API module found and loaded (mots_resolve.py:177)
2022-11-15 11:48:13,738 - StAI - DEBUG:

next line is added 5 minutes later.

Hi

Once speech has been transcribed into text, it should be possible to edit this text directly in the window where it is displayed. After which there should be a button at the bottom of the window: SAVE and TRANSLATE. After pressing the first one, the edited subtitles are saved, after pressing the second one they are translated by Whisper and opened in the window for further editing with the SAVE button. Saving can also be automatic, but will then unnecessarily consume disks in particular SSD.

BaGRoS

Hi, for some reason the time intervals won't accept time past 1 minute.
For example, I can transcribe a timeline set on 30.00 - 40.00. But I can't set anything past 59.00 second.
If I have a portion of timeline I want to transcribe, let's say at 40.00 - 01.23.00 , it wil not work. The terminal windows will say "ValueError: could not convert string to float: '01.23.00'"

Using StoryToolkitAI v0.17.0 on an Intel Mac, OSX 11.4. I installed Homebrew and ffmpeg (if I try to install again terminal says "ffmpeg 5.1.2 is already installed and up-to-date."), but when I launch the app, it alerts me that ffmpeg is not present. ffmpeg is at
/usr/local/Cellar/ffmpeg/5.1.2. Is this as expected?

First of all, thanks for building this! I haven't fully tested it yet but it's already amazing. I hope you keep working on it as well. I have a question regarding the model storage location, especially on Mac. Where can I find it, in case I need to delete files?

Thanks!

I'm using a PC with AMD GPU, and I see that there is no option for Transcribing using my GPU.
Does StoryToolkitAI support Transcribe using GPU (specifically AMD GPU)?

Thanks

The app seems to get stuck when loading the Resolve lib files on some Windows 10 machines, and it's not reporting anything back but simply gets stuck and crashes after a few minutes. This seems to be an issue on both Python 3.9.13 and 3.10.7.

my app.log shows this

2022-12-15 05:17:14,444 - StAI - DEBUG: 
--------------
Platform: Windows 10
 Platform version: 10.0.19044
 OS: 10 10.0.19044 SP0 Multiprocessor Free  ('', '', '')  
 running Python 3.9.13.final.0
-------------- (app.py:140)
2022-12-15 05:17:15,186 - StAI - DEBUG: All package requirements met. (app.py:160)
2022-12-15 05:17:15,187 - StAI - INFO: �[1m�[4mRunning StoryToolkitAI version 0.17.5 (standalone) (app.py:8006)
2022-12-15 05:17:15,683 - StAI - INFO: Config setting ignore_update saved as "False"  (app.py:8130)
2022-12-15 05:17:15,684 - StAI - INFO: Updated config file C:\Users\Rexel Bartolome\StoryToolkitAI\config.json with ignore_update data. (app.py:8168)
2022-12-15 05:17:15,689 - StAI - DEBUG: Looking for ffmpeg in env variable. (app.py:8437)
2022-12-15 05:17:15,689 - StAI - DEBUG: FFMPEG_BINARY env variable is empty. Looking for ffmpeg in PATH. (app.py:8444)
2022-12-15 05:17:15,697 - StAI - DEBUG: Checking ffmpeg binary: C:\ProgramData\chocolatey\bin\ffmpeg.EXE (app.py:8456)
2022-12-15 05:17:16,069 - StAI - DEBUG: FFMPEG exit code: 1 (app.py:8461)
2022-12-15 05:17:16,069 - StAI - INFO: FFMPEG found at C:\ProgramData\chocolatey\bin\ffmpeg.EXE (app.py:8464)
2022-12-15 05:17:16,070 - StAI - INFO: Config setting whisper_model_name saved as "medium"  (app.py:8130)
2022-12-15 05:17:16,071 - StAI - INFO: Updated config file C:\Users\Rexel Bartolome\StoryToolkitAI\config.json with whisper_model_name data. (app.py:8168)
2022-12-15 05:17:16,075 - StAI - INFO: Config setting whisper_device saved as "auto"  (app.py:8130)
2022-12-15 05:17:16,076 - StAI - INFO: Updated config file C:\Users\Rexel Bartolome\StoryToolkitAI\config.json with whisper_device data. (app.py:8168)
2022-12-15 05:17:16,117 - StAI - INFO: Using cuda for Torch / Whisper. (app.py:5947)
2022-12-15 05:17:16,118 - StAI - INFO: Config setting s_semantic_search_model_name saved as "all-MiniLM-L6-v2"  (app.py:8130)
2022-12-15 05:17:16,119 - StAI - INFO: Updated config file C:\Users\Rexel Bartolome\StoryToolkitAI\config.json with s_semantic_search_model_name data. (app.py:8168)
2022-12-15 05:17:16,123 - StAI - INFO: Config setting disable_resolve_api saved as "False"  (app.py:8130)
2022-12-15 05:17:16,124 - StAI - INFO: Updated config file C:\Users\Rexel Bartolome\StoryToolkitAI\config.json with disable_resolve_api data. (app.py:8168)
2022-12-15 05:17:16,124 - StAI - DEBUG: MotsResolve module initialized. (mots_resolve.py:42)
2022-12-15 05:17:16,125 - StAI - DEBUG: Found DaVinci Resolve at the default location: C:\Program Files\Blackmagic Design\DaVinci Resolve\ (mots_resolve.py:142)
2022-12-15 05:17:16,125 - StAI - DEBUG: Unable to find module DaVinciResolveScript from PYTHONPATH - trying default locations next (mots_resolve.py:154)
2022-12-15 05:20:57,752 - StAI - DEBUG: 

Installing it on my F: drive only got up to this point. But the python I have is 3.10 and I can't uninstall it because I need it for using Stable Diffusion, another AI program lol

I installed it on my C: drive inside the program files folder and I also installed python 3.9. When I run it, nothing shows up. When I run as admin, it's the same thing as how it was in my F: drive.

These are my Python installations:

C:\Users\Rexel Bartolome\AppData\Local\Programs\Python\Python39\python.exe
F:\stable-diffusion\Python\Python310\python.exe
C:\Users\Rexel Bartolome\AppData\Local\Microsoft\WindowsApps\python.exe

Just a note to let you know that when I installed StoryToolkitAI v0.16.15 for M1 on my laptop, the install script did not complete successfully until I manually created a 'CommandLineTools' folder in /Library/Developer ... until I did that it would fail on the CommandLineTools install.

Also note that the second instruction brew install ffmpeg would not run until I manually added the necessary PATH variables.

None of it was an issue, but 'install homebrew and ffmpeg with 2 commands" could be untrue/misleading for some users.

Please improve the saving of the rendered audio to the file. After changing from mp4 to mov the files are less well recognised by whisper.
BaGRoS

app.log

I'm not very experienced with ML stuff, so apologies if this is all normal or not within the scope of your extension.

Getting this error when running Transcribe Timeline after the audio render:

UserWarning: FP16 is not supported on CPU; using FP32 instead
  warnings.warn("FP16 is not supported on CPU; using FP32 instead")

Also getting very slow speeds (possibly due to the error?); 200-250 seconds for a 13 second clip.

System specs:
Mac Pro 2019
3.2GHz 16-core Intel Xeon W
192GB RAM
AMD Radeon W6800X Duo 32GB