This repo is intended to be a collection of self contained tools and scripts to make it easy to create custom Nothing Glyph Compositions which in turn can be imported into the Nothing Glyph Composer.
Rick_demo.mp4
- 🎬 Simple tutorial Video
- 📌 Disclaimer
- 🦺 Need help?
- 🚧 Compatibility
- ✏️ Usage
- Hardware limitations
⁉️ Troubleshooting- Locate FFmpeg - Audacity needs the file 'avformat.dll'
- ModuleNotFoundError: No module named 'termcolor'
- I can't find the modified audio file after using GlyphModder
- File is not supported on this device
- File is not created by Composer
- Glyph Composer does not import my file
- No glyphs light up
- I can import my audio but my glyphs don't light up
- Some of my glyphs don't light up
- The Glyphs are not alligned or desync with the audio
- 🔧 The technical details
- Pull requests
https://www.youtube.com/watch?v=YlJBqQxSgWA
These scripts are provided as-is without any warranty. I and all other contributors are not responsible for any damage, misuse or other kind of physical or mental damage which results in the use of these scripts. This repo is in no way shape or form affiliated with Nothing Technology Limited (NOTHING).
If you need help look at the Troubleshooting chapter. If this does not help either feel free to use the Discussion feature on GitHub or reach out to me on Discord: @sebiai
- Linux ✅
- Windows ✅
- Mac ✅
I have only tested the scripts on Linux. Multiple people managed to do it on Windows and MacOS.
- Nothing Phone (1) ✅
- Nothing Phone (2) ✅
Works on both Nothing devices. Supports all 33 Zones on Phone (2).
Note
If you are on Windows you can use (double click) the Install-Dependencies.bat script to install all requirements. If the script issued no warnings or errors you can continue at Making a custom Glyph composition.
-
[Required] python3 - To run the scripts
-
[Required] ffmpeg - To write metadata to the sound file
-
[Required] ffprobe - To read metadata from the sound file (should be included in almost every ffmpeg install)
-
[Optional] Audacity® - Create custom compositions more easily with Labels
➡️ If ffmpeg or ffprobe are not in PATH they can be passed to the script with the
--ffmpegand--ffprobearguments.
git clone https://github.com/SebiAi/custom-nothing-glyph-tools.git
cd custom-nothing-glyph-toolsNote
If you don't feel comfortable using git then you can also download an archive file here.
pip3 install -r requirements.txtSometimes it is also just pip instead of pip3.
I would recommend using Audacity® to make your life easier. You can use it to cut your audio to the right length, layer effects on it, set Labels (which will come in handy later) and more.
You can skip to Creating the Glyph format with Audacity® if you already have the sound cut to the right length and the right codec (opus).
Important
When exporting the sound always use the opus codec:
- Export audio: Files -> Export -> Export Audio...
- Select the right codec in the dropdown: Select Opus - default settings are fine
- Change the extension to ogg when naming your file!
- Export
Important
When exporting your audio with Audacity® a popup might appear asking you to Locate FFmpeg. This is expected on Windows - follow the steps in Locate FFmpeg - Audacity needs the file 'avformat.dll'
To make our lives easier we can utilize Audacity®'s Labels functionality and then use the Label data to generate our two needed csv files with the help of the GlyphTranslator - one for the AUTHOR tag (stores the light data) and the other for the CUSTOM1 (stores data for the app to display) tag (for more info read the technical details).
I would recommend saving your Audacity® project regularly.
Note
You can also use the MidiToLabel script to transform MIDI files to Audacity Labels. Use the --help command on the script to learn more.
- Open your audio with Audacity®
- Add the Labels track: Tracks -> Add New -> Label Track (you can add multiple tracks)
You can now add Labels by clicking on your desired location on the Label Track and pressing CONTROL + b. To remove one Label right-click on the Labels name and choose Delete Label. More info on how to work with them see here.
Each Label should be named like this: [#]glyphId-lightLevelFrom[-lightLevelTo[-Mode]]
- # (Optional): Just the
#symbol to indicate that you want to address one of the 33 individual Zones on Phone (2). - glyphId: See glyphId below.
- lightLevelFrom: In percent 0 to 100.
- lightLevelTo (Optional): In percent 0 to 100. The default is lightLevelFrom.
- Mode (Optional): Currently there are three modes supported (Desmos Graphs):
- LIN: Linear Interpolation (default)
- EXP: Exponential Interpolation
- LOG: Logarithmic Interpolation
Important
The brackets ([ and ]) mean optional. Therefore do NOT include them in the Label name!
If you like regex patterns, the name of the Label should match this one (thanks Joel05):
^((?:[1-9]|1[0-1])|(?:#(?:[1-9]|[1-2]\d|3[0-3])))-(\d{1,2}|100)(?:-(\d{1,2}|100))?(?:-(EXP|LIN|LOG))?$Important
At the end of the audio there MUST be a non ranged Label called END. This is needed so the script knows how long the audio is. Also, make sure that there is enough space between your last Glyph lighting up and the END Label or else it might not get played - add silence at the end of the audio file if that happens and move the END Label accordingly.
Depending on which Phone you want to make the composition for there are two modes that I have called the Compatibility or Phone2 mode:
After you've made your decision continue to the respective section:
- You can NOT use the
#symbol in front which is used to control one of the 33 Zones of Phone (2). - The
glyphIdcan only be 1 to 5 which correspond to the individual Glyphs:
| glyphId | Glyph |
|---|---|
| 1 | Camera |
| 2 | Diagonal |
| 3 | Battery |
| 4 | USB Line |
| 5 | USB Dot |
Example c1:
0.000000 1.000000 2-0-100-LIN
1.000000 2.000000 2-100-0-LIN
2.000000 3.000000 3-0-100-LIN
3.000000 4.000000 3-100-0-LIN
5.000000 5.000000 END
In this example, the Mode -LIN is not necessary because it is the default but makes it more readable.
Example c2:
0.000000 1.000000 2-0-100
1.000000 2.000000 2-100-0
2.000000 3.000000 3-0-100
3.000000 4.000000 3-100-0
5.000000 5.000000 END
This is a minimalized version of Example c1.
Example c3:
0.000000 0.500000 2-100
0.000000 1.000000 3-100-0
0.500000 1.500000 4-0-100-EXP
2.000000 2.000000 END
Another example with one exponential interpolation (-EXP) and line one and two are minimalized. They expand to 2-100-100-LIN and 3-100-0-LIN.
Please continue at the Exporting Labels section.
The glyphId can only be 1 to 11 which correspond to the individual Glyphs:
| glyphId | Glyph |
|---|---|
| 1 | Camera top |
| 2 | Camera bottom |
| 3 | Diagonal |
| 4 | Battery top right |
| 5 | Battery top left |
| 6 | Battery top vertical (left side) |
| 7 | Battery bottom left |
| 8 | Battery bottom right |
| 9 | Battery bottom vertical (right side) |
| 10 | USB Line |
| 11 | USB Dot |
Important
Do NOT prepend the # symbol! This is for addressing each individual Zone - see below.
If you want even more control you can control each individual Zone with the Zone ids:
| Index | Glyph | Direction |
|---|---|---|
| 1 | Camera top | - |
| 2 | Camera bottom | - |
| 3 | Diagonal | - |
| 4-19 | Battery top right | From right to left |
| 20 | Battery top left | - |
| 21 | Battery top vertical (left side) | - |
| 22 | Battery bottom left | - |
| 23 | Battery bottom right | - |
| 24 | Battery bottom vertical (right side) | - |
| 25 | USB Dot | - |
| 26-33 | USB Line | From bottom to top |
Example p1:
1.000000 2.000000 1-100
2.500000 3.500000 2-100
4.000000 5.000000 3-100
5.500000 6.500000 4-100
7.000000 8.000000 5-100
8.500000 9.500000 6-100
10.000000 11.000000 7-100
11.500000 12.500000 8-100
13.000000 14.000000 9-100
14.500000 15.500000 10-100
16.000000 17.000000 11-100
18.000000 18.000000 END
Go through each of the 11 Glyphs where each Glyph is on for 1s and with 0.5s delay in between.
Example p2:
1.000000 2.000000 #1-100
2.500000 3.500000 #2-100
4.000000 5.000000 #3-100
5.500000 6.500000 #4-100
7.000000 8.000000 #5-100
8.500000 9.500000 #6-100
10.000000 11.000000 #7-100
11.500000 12.500000 #8-100
13.000000 14.000000 #9-100
14.500000 15.500000 #10-100
16.000000 17.000000 #11-100
17.500000 18.500000 #12-100
19.000000 20.000000 #13-100
20.500000 21.500000 #14-100
22.000000 23.000000 #15-100
23.500000 24.500000 #16-100
25.000000 26.000000 #17-100
26.500000 27.500000 #18-100
28.000000 29.000000 #19-100
29.500000 30.500000 #20-100
31.000000 32.000000 #21-100
32.500000 33.500000 #22-100
34.000000 35.000000 #23-100
35.500000 36.500000 #24-100
37.000000 38.000000 #25-100
38.500000 39.500000 #26-100
40.000000 41.000000 #27-100
41.500000 42.500000 #28-100
43.000000 44.000000 #29-100
44.500000 45.500000 #30-100
46.000000 47.000000 #31-100
47.500000 48.500000 #32-100
49.000000 50.000000 #33-100
51.000000 51.000000 END
Same as Example p1 except for lighting up all the Glpyhs we control the individual Zones.
Looking at the examples of the Compatibility mode can also help.
Please continue at the Exporting Labels section.
- File -> Export -> Export Labels
Now that we have a Labels file we can use the GlyphTranslator to get our desired files like this:
python3 GlyphTranslator.py MyLabelFile.txtNote
You can also add a watermark by writing your watermark to a text file and passing it to the script like this: --watermark MyWatermarkFile.txt
Assuming your Label file was called MyLabelFile.txt it will spit out two files called MyLabelFile.glypha and MyLabelFile.glyphc1 in your current working directory.
Important
If you see this message in the output:
INFO: Auto detected Phone (1) and Phone (2) compatibility mode.
INFO: If you intended to use the Glyphs 1-5 on the Nothing Phone (2) use the '--disableCompatibility' parameter. More info with '--help' or in the README.
act according to this:
- You are using Compatibility mode: Do NOT follow the advice - this is intended behaviour.
- You are using Phone2 mode: Please add the
--disableCompatibilityparameter to the command to use the right Glyphs.
When you have both your .glypha and .glyphc1 files (via Audacity® and the GlyphTranslator method or otherwise) you can write them to an audio file as metadata:
python3 GlyphModder.py -t MyCustomTitle -w MyLabelFile.glypha MyLabelFile.glyphc1 MyGlyphCreation.oggNote
The -t argument is optional, this just sets the TITLE tag which, as far as I could see, is not used anywhere in the Glyph Composer right now.
Note
You can provide the path to ffmpeg with the --ffmpeg argument if it can not be found in PATH.
Important
Regarding the -w option: The .glypha file must be passed before the .glyphc1 file. The script will still comply but you can not play the file in the Glyph Composer!
Congrats, you can now transfer the audio file to your Nothing phone and import it into the Glyph Composer app!
This is used when you want to modify a composition but you don't have the Audacity®/Label file.
python3 GlyphModder.py MyGlyphCreation.oggAssuming your audio file was called MyGlyphCreation.ogg it will spit out two files called MyGlyphCreation.glypha and MyGlyphCreation.glyphc1 in your current working directory.
Note
You can provide the path to ffprobe with the --ffprobe argument if it can not be found in PATH.
It is almost impossible to convert the .glypha and .glyphc1 files back to an Audacity® Label file therefore you would need to edit both files manually. More info on that is in the The technical details section.
Note
The desync between the audio and the light sequence when the composition is set as a ringtone or notification sound seems to be an OS issue and only seems to occure on compositions longer than 10 seconds. No fix for this issue is in sight in the near future.
- The Glyphs can't playback fast-changing light sequences. One user reported that this phenomenon disappeared when it was set as a ringtone or notification sound. When doing this there is a small chance that the audio and the light sequence get desynced over time. The only solution until now is to don't make fast light sequences.
- In some unknown cases the light sequence gets desynced from the audio when set as a ringtone or notification sound.
This error is expected on Windows and can be solved the following way:
- Download and extract this file to a location you'll remember (you might need 7-Zip to extract it): https://www.gyan.dev/ffmpeg/builds/ffmpeg-release-full-shared.7z
- Then click on the Browse... button in Audacity® and navigate to the extracted content
- Navigate into the
binfolder - Select the
avformat.dll- The filename might differ a little bit - Click OK in the Audacity® dialog
You did not properly install the requirements. Try executing
python3 -m pip install -r requirements.txtThe file was modified in place, no other files were generated.
If you want to confirm that the metadata was written correctly see Glyph Composer does not import my file.
If you get this error message:
Import failed. File is not supported on this device.
You are trying to import a Phone (2) exclusive (Phone2 mode) composition on a Phone (1).
If you created this composition please only use glyphIds that are compatible with your device. See glyphId.
If you get this error message:
Import failed. File is not created by Composer.
- Avoid using a third party file explorer app to import the audio file and use Android's built in instead. Apps like Solid Explorer may try to modify the file and the import will fail.
- You are trying to import an audio file with no metadata. See No or missing metadata.
Make sure that the file has the right codec and that the metadata is present:
ffprobe MyGlyphCreation.oggIt should spit out something like this:
Input #0, ogg, from 'MyGlyphCreation.ogg':
Duration: 00:00:10.01, start: 0.000000, bitrate: 146 kb/s
Stream #0:0: Audio: opus, 48000 Hz, stereo, fltp
Metadata:
encoder : Lavf58.76.100
TITLE : MyCustomSong
ALBUM : CUSTOM
AUTHOR : eNrtxqENAAAIAzBPwieI/X8dhg+wTU0zOV1xd3d3d3d3d3d3d3d3d/f3F8ocpN8=
COMPOSER : Spacewar Glyph Composer
CUSTOM1 : eNoljVsKAEEIwy60A+r4qPe/2E7pj8FAiZ340vvYh8S7HjBiPB7ivUQ1ZexS3owkUJxlDGWOUZZf
: yqrmrs24awVahVFhVIAKUAEqrAqrwg8LSR98
Important are:
- Audio: opus
- The extension
.ogg - The presence of the metadata tags TITLE, ALBUM, AUTHOR, COMPOSITOR and CUSTOM1 (the order is irrelevant) - Optionally the CUSTOM2 tag if the composition is for Phone (2).
You have two options:
- Reexport with Audacity® (see Cutting the sound)
- Convert with ffmpeg (replace
MyGlyphCreation.oggwith your audio):ffmpeg -i MyGlyphCreation.ogg -strict -2 -c:a opus -map_metadata 0:s:a:0 output.ogg
See Wrong codec
Did you run the GlyphModder correctly? See Write to an audio file.
See Glyph Composer does not import my file.
When you use the linear, exponential or logarithmic mode make sure that your Labels cover a range like this left Label:
You can make the right Label ranged if you drag on the left or right part of it.
See I can import my audio but my glyphs don't light up.
Click to expand
This is just a written up more technical view on this whole system.
The audio seems disconnected from the lighting - seems like it only determines the final length of the composition. The maximum you can produce in the app is 10s but it can be longer when making it custom.
There are the following tags in the ogg file (order irrelevant):
- TITLE
- ALBUM
- AUTHOR
- COMPOSER
- CUSTOM1
- CUSTOM2
The AUTHOR and CUSTOM1 tags contain both Base64 encoded and then zlib compressed data (Best Compression (no preset dictionary) - see here).
Contains the title given in the Glyph composer. Weirdly enough the Glyph Composer does not use this tag when displaying the name of the composition - the filename is used instead.
Saves what sound pack was used when the composition was created. Can be changed without any effect on the audio or lights (could affect the preview in the Glyph Composer). It does display in the Glyph Composer.
After decoding and decompressing it contains the Glyph light data in a csv like manner where in each line we have the 5 Glyphs/33 Zones separated and followed by a comma (,). Each line corresponds to 16ms.
Depending on if the CUSTOM2 tag is set to 33cols we have, what I will call, the Compatibility or Phone2 mode:
- Not set => Compatibility mode - 5 columns
- Set => Phone2 mode - 33 columms (Nothing Phone (2) exclusive)
Depending on the mode we have different indexes in the csv for the Glyphs:
| Index | Glyph |
|---|---|
| 0 | Camera |
| 1 | Diagonal |
| 2 | Battery |
| 3 | USB Line |
| 4 | USB Dot |
| Index | Glyph | Direction |
|---|---|---|
| 0 | Camera top | - |
| 1 | Camera bottom | - |
| 2 | Diagonal | - |
| 3-18 | Battery top right | From right to left |
| 19 | Battery top left | - |
| 20 | Battery top vertical (left side) | - |
| 21 | Battery bottom left | - |
| 22 | Battery bottom right | - |
| 23 | Battery bottom vertical (right side) | - |
| 24 | USB Dot | - |
| 25-32 | USB Line | From bottom to top |
Each Glyph can have a light value from 0 to 4080 and it appears that the smallest step is 1. If the data is longer than the audio it will not be played.
The new line consists of Carriage Return (CR) and Line Feed (LF): \r\n
The data ends with a final new line \r\n.
Most of the data is padded at the end with multiple "zero lines" (0,0,0,0,0,) even if the audio track has ended - therefore also the playback of the light show.
This might be because of how the light data is stored in the app itself which are csv files for every audio clip which are combined.
Example 5 Glyphs (Compatibility mode):
0,0,4080,0,0,
0,0,4080,0,2032,
0,0,0,0,0,
0,0,0,0,0,
0,0,0,0,0,
The Battery Glyph is fully on for 32ms and the USB Dot Glyph is only on for 16ms after 16ms (after the start) at about 50% brightness.
Example 33 Zones (Phone2 mode):
0,0,0,2709,2709,2709,2709,2709,2709,2709,2709,2709,2709,2709,2709,2709,2709,2709,2709,0,0,2709,0,0,0,0,0,0,0,0,0,0,0,
0,0,0,1791,1791,1791,1791,1791,1791,1791,1791,1791,1791,1791,1791,1791,1791,1791,1791,0,0,1791,0,0,0,0,0,0,0,0,0,0,0,
0,0,0,919,919,919,919,919,919,919,919,919,919,919,919,919,919,919,919,0,0,919,0,0,0,0,0,0,0,0,0,0,0,
0,0,0,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,245,0,0,245,0,0,0,0,0,0,0,0,0,0,0,
This depends on what device the composition was created on:
- Phone (1):
Spacewar Glyph Composer - Phone (2):
Pong Glyph Composer
If this does not match the Glyph Composer will not import the file.
It is totally possible to import a Phone (2) composition IF the CUSTOM2 tag is NOT set, which infers that no 33 Zone sound pack was used.
This is mainly data for the Glyph Composer so it can display the timeline when playing the file. After decoding and decompressing each dot in the app is defined by a timestamp (in ms and 1 ms steps) and a Glyph id (see AUTHOR) separated by a dash (-). Between each dot is a comma (,) and at the end of the line also.
There are no new lines in this file, all dots are after one another. It is entirely possible to mismatch the CUSTOM1 and ALBUM data (could affect the preview in the Glyph Composer).
Keep in mind that the Glyph Composer only has 5 Buttons (therefore 5 dot locations) which means that complex custom light animations which address Zones individually can not be properly displayed in the Glyph Composer.
This indicates if the saved data in the AUTHOR tag uses the 33 Zone addressing (Phone2 mode) instead of the 5 Glyphs (Compatibility mode) addressing.
This tag will only be present if the composition was made on a Nothing Phone (2) and with a 33 Zone sound pack (e.g.: Swedish House Mafia) therefore this composition can only be played back on a Nothing Phone (2).
If it is present it ALWAYS has the value 33cols.
It can be manually imported on a Nothing Phone (1) by moving the audio file to Ringtones/Compositions but the phone will interpret the lighting data in the Compatibility mode (see Author tag).
Pull requests are welcome (improvements, new scripts/tools).