SubStitcher

Opus chaptered audiobook player, encoder and editor with colored subtitles and fancy fonts. Transcribe to vtt subtitles, search through all Chapters, History, Playlist, Bookmarks, Fonts, Colors, Words, Subs, Stats. Dictionary word lookup. Basic Video Editing. Mac (dmg or homebrew), Linux (appimage, AUR, flatpak), Windows (zip), Android

Encoding opus Chaptered Audiobooks

• Encode 16kbps audiobooks which is 4x/6x/8x smaller than 64/96/128kbps
• 32kpbs only use for audio like Quran recitations and it's about 2.5x larger in file size due to vbr
• opus (2012) is a superior audio codec compared to mp3 (1993) or aac (1997) at low bitrates
• max 100 hours and 999 chapters per audiobook
• if exceed limits will offer to automatically split into multiple audiobooks showing the split points
• Title Case chapter titles and regular expression replace
• Remove silence -26dB, -30dB, -34dB, -38dB, -42dB, -46dB, reduces transcribing hallucinations
• Hiss (reduction) preview a random audio to compare
• Batch Trim Audio beginning and end, previews 6 trimmed audios
• Extract chapters with names from audiobooks
• Edit Metadata (chapters, author, title) of opus audiobook
• Paste List of text to rename files, two surah lists
• denoise audio with DeepFilterNet3 (36MB for runtime and model). Old audio with static, hiss, keyboard clicking, fan noise removes all that noise and leaves just the speech. ~22x realtime speed on Mac M1.

Transcribe

• Transcribe with 30 second segments to reduce hallucination with whisper.cpp, keeps model in memory for entire chapter
• Repeats vtt to remove repeated words, capitalize pronouns, Islamic terms and honorifics
• Auto-detect language
• Automatically translate to English only from a foreign language if using whisper 2 models but not turbo 3
• Custom Prompt should use language of audio being transcribed. Use a few punctuation. Remebers last 4 used prompts plus default.

Transcribe for the following languages:

• Afrikaans, Albanian, Amharic, Arabic, Armenian, Azerbaijani, Basque, Belarusian, Bengali, Bosnian, Bulgarian, Catalan, Cebuano, Chichewa, Chinese, Cantonese (CN), Cantonese (HK), Mandarin (TW), Corsican, Croatian, Czech, Danish, Dutch, English, Esperanto, Estonian, Filipino, Finnish, French, Western Frisian, Galician, Georgian, German, Greek, Gujarati, Haitian Creole, Hausa, Hawaiian, Hebrew, Hindi, Hmong, Hungarian, Icelandic, Igbo, Indonesian, Irish, Italian, Japanese, Javanese, Kannada, Kazakh, Khmer, Kinyarwanda, Korean, Kurdish, Kyrgyz, Lao, Latin, Latvian, Lithuanian, Luxembourgish, Macedonian, Malagasy, Malay, Malayalam, Maltese, Māori, Marathi, Mongolian, Myanmar, Nepali, Norwegian, Odia, Pashto, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Samoan, Scottish Gaelic, Serbian, Sesotho, Shona, Sindhi, Sinhala, Slovak, Slovenian, Somali, Spanish, Sundanese, Swahili, Swedish, Tajik, Tamil, Tatar, Telugu, Thai, Turkish, Turkmen, Ukrainian, Urdu, Uyghur, Uzbek, Vietnamese, Welsh, Xhosa, Yiddish, Yoruba, Zulu

Playback

• set playback speed 0.5x to 2.0x
• chapter title hide/show for anki audiobooks
• copy metadata audiobook author & title file size
• copy chapter list and write to text file

Subtitles

• vtt subtitles if srt converts automatically to vtt
• set default font, font size, font line spacing
• apply default font, font size, font line spacing
• left/right arrows advanced to prev/next subtitle
• use primary / secondary at same time, i.e. bilingual subs
• copy to clipboard primary or secondary subtitle line (u / U)
• copy to clipboard in memory (font conversion for primary)
• search foreign language subtitles
• blur shadow (uses 2x cpu usage) but dims it a bit
• font size ↑/↓ or Alt⌥+↑/↓ ...use later shortcuts when Font/Colors panel is open

Dictionary Lookup Words

• d opens dictionary word overlay
• click on word to open Apple Dictionary on Mac
• or on Windows, Linux, Android copy to clipboard
• filters out tons of common words (and, the, to, from, etc.)
• removes repeated words
• History of looked up words clicked on
• Clear or Copy All history to clipboard

• Pause mode 2s, 3s, 5s, 10s or Dictionary mode (forever)
• CJK (Chinese, Japanese, Korean) and Arabic subtitles
• tokenizes Japanese with tiny segmenter
• Chinese and Korean tokenizes pretty well
• Apple Dictionary has following languages to English available as of February 2026:
• Arabic, Bangla, Cantonese, Simplified Chinese, Traditional Chinese, Croatian, Czech, Danish, Dutch, Finnish, French, German, Greek, Gujarati, Hindi, Hungarian, Indonesian, Italian, Japanese, Kannada, Kazakh, Korean, Malay, Malayalam, Norwegian, Polish, Portuguese, Punjabi, Russian, Slovak, Spanish, Swedish, Tamil, Telugu, Thai, Turkish, Ukranian, Urdu, Vietnamese

Subtitle Manager

• bilingual subtitles
• audiobookname_vtt sub directory to keep many subs (autoloads)
• automatically changes font size based on length of subtitles
• v Subtitle Manager for bilingual subtitles and loading subs
• x swap top and bottom subtitles
• set font and color for 'active subtitle' which is the bottom one
• set font and color for each one top and bottom
• converts srt to vtt automatically

Sleep Timer

• set to 15, 30, 45, 60, 90, 120 minutes
• z sleep at chapter end
• sleep at End of Audiobook
• Z cancel sleep timer
• pause or adjusting playback speed also cancels sleep timer
• warning of 120 seconds before closing app
• if paused when setting sleep timer, playback is started
• set to Pause instead of Close app so translating or transcribing continues while you sleep

Chapters

• search and prevent chapters from players based on chapter title
• shuffle chapters for lectures or vocab learning
• keeps track of which chapters have been shuffled so doesn't repeat
• shows duration of each chapter (ffprobe)
• search through all chapters in entire plays (under Subs panel)

History

• shows audiobook title with chapter title and time position
• duration of audiobook
• percent progress of total timeline position
• timeline position of where it'll resume
• press h 1-9 to quickly open history entries
• sorted by most recent
• only shows most recent chapter with a particular audiobook

Playlist

• set up to 10 playlists and switch between them
• search through entire playlist
• p 1-9 to quickly open playlist entries

Bookmarks

• shows audiobook title with chapter title and time position
• pin up to 9 bookmarks
• press b 1-9 to quickly open bookmark entries
• mutiple bookmarks from same audiobook grouped together
• sorted by most recently added

Fonts

• ligatures fonts, missing ligatures, alternates, fonts must be uppercase
• 100+ fonts
• demo fonts
• free (free for commerical use)
• each missing and each alternate font, subs must be converted
• ligature demo fonts, subs need converting only once
• alternates don't work with mixed CJK and Latin text on same subtitle line
• favorites list
• choose own custom folder for own fonts, refresh for added fonts, must restart app though to clear out removed fonts
• custom folder can have subfolders although all fonts will be sorted alphabetically
• 500+ LUTs (color presets) applied to subtitles and existing color palettes, hidden to right of Stats tab

• download 38 demo fonts or link is provided for each one within app
• custom fonts can all be in one directory within various subdirectories, it recursively loads the fonts
• 
• in font panel set custom folder to this one and all fonts will be loaded

│  customfonts
├── alternates
│   ├── Aloevera.ttf
│   ├── Bentley Vintage.otf
│   ├── Kambegi.ttf
│   └── Mount Hills.otf
├── ligatures177
│   ├── Brilliant Heavens demo.otf
│   ├── Categories Elegant demo.otf
│   ├── Changing Campaign demo.otf
│   ├── Creating Families demo.otf
│   ├── Creating Graphics demo.otf
│   ├── Enduring demo.otf
│   ├── Gares demo.otf
│   ├── Reminder According demo.otf
│   ├── Roommate Surrealism demo.otf
│   ├── Salvador Abstract demo.otf
│   └── Traditional Civilization demo.otf
├── ligaturesVarious
│   ├── Chocolate Chips.otf
│   ├── Pricedown Black.otf
│   ├── Rocket Raccoon free.otf
│   ├── Shoese Flower.ttf
│   ├── Sophia Melanie.otf
│   └── Souther Daleska demo version.otf
├── missingligatures177
│   ├── Abstract Settings demo.otf
│   ├── Classical Aesthetics demo.otf
│   ├── Coastline Classical demo.otf
│   ├── Created Aesthetic demo.otf
│   ├── Creates Presence demo.otf
│   ├── Engaging Realities demo.otf
│   ├── Fondness Romance demo.otf
│   ├── Healthcare Resilience demo.otf
│   ├── Radical Blending demo.otf
│   ├── Realities Endlessly demo.otf
│   ├── Rococo Aesthetic demo.otf
│   └── Titanium Galleries demo.otf
├── mustBeUppercase
│   ├── Children Interests demo.otf
│   └── Intricate Narrative demo.otf
└── seesawcase
    ├── Harquil.ttf
    ├── Sidethree.otf
    └── Zigzageo.ttf

Glyph Viewer

only 1-9, basic latin and ligatures

Colors

• 750+ color palettes
• choose between coloring words or letters
• 20 and 12 colors per palette
• 91 same (similar) Palettes (font, stroke, shadow)
• monochromatic, food palettes
• letters work with alternates but not with ligatures
• favorites list, shortcut key R (rang रंग Hindi is color)

Words

• analyzing vtt subtitles for word frequencies
• top 500 words by frequency
• click top 500 words to search subtitles and paragraphs
• top 7-word, 6-word, 5-word, 4-word, 3-word phrases
• click phrases to do exact phrase match in subs and paragraphs
• words are automatically analyzed after 20 seconds of audiobook loading

Subs

• search through entire subs and click on results to navigate to time position in audiobook
• search paragraphs too, paragraphs are ~ 8 sentences
• click paragraph to copy to clipboard
• search through entire playlist chapters of entire playlist
• indexes all chapters for every audiobook to be searched
• press ENTER to search subtitles
• / to start search (focus in search field)
• Ctrl+Backspace or Cmd+Backspace to clear search results
• TAB to focus elsewhere so keyboard shortcuts work

Stats

• Shows Active Days which are at least 30 mins of listening
• Daily streaks
• Longest 3 days of total listening time
• Today, Yesterday, 2 to 10 days ago of listening activity
• shows duration of listen audiobook title as well as which chapter listened to
• shows total cumulative listening time of each audiobook from all chapters
• average time per chapter
• top 50 audiobooks listened to by duration
• time duration bars of listening time for last 30 active days
• stats are stored in ~/.config/substitcher/watch_tracking_time

Slicing / Editing Audio

• make cuts (slices of audio) in subdirectory audiobookname_cuts
• make an audiobook from audiobookname_cuts and transcribe for subs
• first encode video to audiobook then slice on audiobook, avoids keyframe issues for precise timing
• i and o to set in/out point
• Listen plays 900ms of audio and pauses
• backward, Listen to audio from position
• foward, Listen to audio just before position
• ; Listen, seek to End of Sub
• j and k Listen, backward/forward 1s
• , and . Listen, backward/forward 100ms

Adhan Clock

• prayer times
• White Days
• auto-detect location or set coordinates
• location is valid for 30 days (cached)

Anki to Opus

• Anki to opus chaptered audiobook (4x repeat vocab, show 2x front, 2x back subs)
• if over 999 notes (rows with audio) then automatically creates multiple audiobooks part 1, part 2, part 3, etcetera
• Arabic, Bengali, Chinese, Czech, Danish, Dutch, English, Filipino, French, German, Greek, Hebrew, Hindi, Hungarian, Indonesian, Italian, Japanese, Korean, Marathi, Norwegian, Pashto, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Spanish, Swedish, Tamil, Telugu, Thai, Turkish, Urdu, Vietnamese
• Nearly 1,000 anki to opus audiobooks on telegram channel https://t.me/Anki2Opus if too lazy to make your own

sample anki to opus audiobook from here 124MiB
down to 13MiB with 4x repeating audio 1h 57m

try Pause mode 2s, Hide Chapter Title, Shuffle

Translate vtt Subtitles

• translate with TranslateGemma
• choose on vtt file and translation into 16 different langauges simutaneously
• using translategemma-4b-it.Q4_K_M.gguf 2.5GB model, works with only 8GB RAM
• translate into these languages (either from or to)
• English, Arabic, Dutch, French, German, Italian, Portuguese, Spanish, Swedish, Russian, Chinese, Japanese, Korean, Thai, Vietnamese, Indonesian, Bengali
• missing Filipino, Hindi, Polish, Turkish due to bad accuracy
• resume translation if necessary, choose same vtt again
• pause and/or unload model (to free up memory) then resume
• unload model becomes avaialble once current subtitle cue (line) is translatedresume and unload model
• online vtt subs translate much faster but vvt must be 977KB or less
• this is offline and subs can be 7MB for instance
• once translating subs is done it'll copy original subtitle into filename_vtt and create a filename_vtt.zip file with all subs
• so all one needs is filename.opus and filename_vtt.zip
• SubStitcher automatically can read all subtitles in filename_vtt subdir in Subtitle Manager

Youtube

• handles videos or audio to stream (ignored for History, Stats), disabled on android/iOS
• download video or audio and playlists with option to resume includes soundcloud and spreaker
• displays subs automatically if avaiable based on default language
• choose audio streams
• choose up to 10 languages to prompt when default language isn't found, this enables a shortlist rather than scrolling through 79 languages each time

list of languages

• English, Afrikaans, Albanian (Shqip), Amharic (አማርኛ), Arabic (العربية), Armenian (Հայերեն), Azerbaijani (Azərbaycan), Belarusian (Беларуская), Bengali (বাংলা), Bhojpuri (भोजपुरी), Bosnian (Bosanski), Bulgarian (Български), Burmese (မြန်မာ), Catalan (Català), Chinese - Simplified (简体), Chinese - Traditional (繁體), Chinese - Cantonese (粵語), Croatian (Hrvatski), Czech (Čeština), Danish (Dansk), Dutch (Nederlands), Estonian (Eesti), Filipino (Tagalog), Finnish (Suomi), French (Français), Georgian (ქართული), German (Deutsch), Greek (Ελληνικά), Gujarati (ગુજરાતી), Hausa (هَرْشٜىٰن هَوْسَا), Hebrew (עברית), Hebrew (עברית), Hindi (हिन्दी), Hungarian (Magyar), Icelandic (Íslenska), Indonesian (Bahasa Indonesia), Italian (Italiano), Japanese (日本語), Javanese (Basa Jawa), Kannada (ಕನ್ನಡ), Kazakh (Қазақ тілі), Korean (한국어), Kyrgyz (Кыргызча), Lao (ລາວ), Latvian (Latviešu), Lithuanian (Lietuvių), Macedonian (Македонски), Malay (Bahasa Melayu), Malayalam (മലയാളം), Maltese (Malti), Marathi (मराठी), Mongolian (Монгол), Nepali (नेपाली), Norwegian (Norsk bokmål), Persian (فارسی), Polish (Polski), Portuguese (Português), Portuguese - Brazil (Português Brasil), Portuguese - Portugal (Português Portugal), Punjabi (ਪੰਜਾਬੀ), Romanian (Română), Russian (Русский), Serbian (Српски), Slovak (Slovenčina), Slovenian (Slovenščina), Spanish (Español), Swahili (Kiswahili), Swedish (Svenska), Tajik (Тоҷикӣ), Tamil (தமிழ்), Telugu (తెలుగు), Thai (ไทย), Turkish (Türkçe), Turkmen (Türkmençe), Ukrainian (Українська), Urdu (اردو), Uyghur (ئۇيغۇرچە), Uzbek (Oʻzbekcha), Vietnamese (Tiếng Việt)

Video Editing (not on Android)

• basic video editing to make multiple cuts and combine cuts
• cuts in Substitcher use hardware accelerated cuts which take longer but are precise cuts
• cuts are huge in file size but combine all cuts encodes to x264 or x265 which are cpu encoded and small in file size
• LosslessCut works great but can't make accurate cuts
• blur up to two regions, only for cuts
• even though ffmpeg is bundled with SubStitcher it is limited to audio only pretty much due to licensing restrictions since x264 and x265 are GPL so if wish to use Video Editing must install the full ffmpeg
• flutter limitations with media_kit don't allow filters to be applied in realtime unlike mpv lua scripts. So can't have audio filtering nor LUTs showing on video
• HOME seek to start of media, END seek to end of media

Motion-Tracking Blur (MacOS only)

• motion-tracking blur as well as invert mode which blurs out everything except motion-tracked object
• motion-tracking has to encode cut once, run motion-tracking for object on cut, then encode cut again, so be patient
• considering motion-tracking blur on Windows (Linux not possible)

LUTs

• 500+ LUTS (lookup tables which are color presets for images/videos)
• all free and can be applied to video cuts
• Lut Picker showing original set in point frame with 5 LUTs previewed at a time
• choose category/author for quick navigation
• favorites list
• 200MB cache limit and clears on app close
• Preview all 500+ LUTS quickly use html thumbnail gallery
• 

Installation

macOS 11.0+ (arm64 Silicon m1,m2,m3,m4,m5) dmg or homebrew

releases or via homebrew in Terminal do

Homebrew

brew install --cask mrfragger/substitcher/substitcher

brew update
brew info substitcher
brew upgrade substitcher
brew uninstall substitcher

after installing dmg from releases or via homebrew in terminal remove quarantine flag

xattr -dr com.apple.quarantine /Applications/SubStitcher.app

Windows x64

just unzip SubStitcher-windows-x64.zip releases
if nothing appears, may need to install Visual C++ 2015-2022 Redistributable

Linux Appimage, AUR, Flatpak

Linux Appimage
appimage right click on file and choose Properties, then Permissions
check Allow executing file as program
or do below

chmod +x  substitcher-x64.AppImage

Transcribing with whisper.cpp won't work on CPUs without AVX which is usually ones before 2014

Linux Arch AUR package yay or paru

yay -S substitcher-bin

paru -S substitcher-bin

Flatpak

flatpak install substitcher-x64.flatpak

Android

Android (64-bit devices only, arm64-v8a)

• Requires Android 7.0+ (API 24+)
• ARM64 devices (99% of modern Android phones, 2015+)
• Tap Settings when prompted and enable installation from your browser

iOS (may publish in future)

• with subs (just one color, one font) nPlayer $5
• no subs for opus audiobooks, vlc (set audio to resume)

Various

Opus audio codec

• outperforms MP3 and AAC at very low bitrates (~16 kb/s)

• Hybrid Architecture SILK (linear predictive coding) for speech-like signals.

• Low Algorithmic Delay and Frame Flexibility Supports frame sizes from 2.5 ms up to 60 ms, allowing very low latency if needed. Fine control of bitrate and delay trade-offs further improves coding of speech/music at low rates

• -application voip gives best quality at a given bitrate for voice signals. It enhances the input signal by high-pass filtering and emphasizing formants and harmonics

Miscellaneous

• Substicher is a flutter app so small in size. It appears big but isn't all that huge.
• 29MB fonts, 65MB LUTs, 36MB DeepFilterNet3 (denoise), 17MB adhan.
• So Mac app size is 232MB - 147MB (assets) = 85MB real size of flutter app
• Never will get a light theme nor support music
• Never will support cover images, choose audiobooks with covers and not to use subtitles, use kid3 app (qt free cross-platform app) for embedding a cover image
• Reason is most audiobook players don't support subtitles, and ones that do, do so due to video support and having a cover image in background intefers with subtitles in most cases
• It just puts a 16x9 black png image with META_BLOCK_PICTURE some info which is conforms to vorbis comment specification
• encode wma or flv or other old codecs just batch convert them to opus then encode those to an opus chaptered audiobook
• parallel ffmpeg -i {} -c:a libopus -b:a 32k {.}.opus ::: *.wma
• parallel ffmpeg -i {} -c:a libopus -b:a 32k {.}.opus ::: *.flv
• Developed since 2022 but in Dec 2025 decided to port from mpv front-end lua scripts with uosc ui, LUTs to a flutter / dart app (not just Mac/Linux anymore) and make it purely an audiobook player with some basic video editing
• Edit fonts, create own fonts https://www.glyphrstudio.com/app (free)
• view detailed info about fonts https://fontdrop.info (free)

Mac App Recommendations

• Transnomino, free file-renamer https://www.transnomino.com
• Lulu, free open-source firewall https://objective-see.org/products/lulu.html
• Stats, free menu bar stats monitoring https://github.com/exelban/stats
• Mole, for terminal, cleans up disk space https://github.com/tw93/Mole
• KeepassXC, free password manager https://keepassxc.org
• Screen Kite, free record screen area with system sound, better than OBS https://www.screenkite.com
• Denoise DeepFilterNet3 in realtime on mic input https://github.com/Ghostkwebb/MetalVoice
• VSCodium, half file size of VSCode https://vscodium.com
• Text Power Tools, extension for VSCodium, powerful text manipulation
• LocalSend, flutter app share files, free https://localsend.org
• Shotcut, add images, mutli-track video editor, free, learning curve https://shotcut.org
• DaVinci resolve free up to 1080p but requires 16GB RAM bare minimum and massive learning curve, expensive

tts (text to speech) that support chapters

my opinon of voice quality

• kokoro > edge > piper
• Qwen3 and echo-tts are better than kokoro but require 16GB RAM minimum
• KokoroTTS, piper.ttstool.com and edge-tts all are free to use

https://github.com/kjyv/KokoroTTS Mac only

• English (American and British) only
• stays at 1GB memory usage amazingly
• Paste text, save wav audio file
• Just as fast as mlx-audio without RAM hitting 10GB due to python

https://piper.ttstool.com

• ~63MB download per voice, fast local cpu processing
• (~3min mac m1 for 30KB text), right-click save as wav audio file
• English (Great Britain)
• alba[medium]
• northern_english_male[medium]
• English (United States)
• amy[medium]
• hfc_female[medium]
• hfc_male[medium]
• joe[medium]

https://github.com/rany2/edge-tts 10.2k
edge-tts --voice en-CA-LiamNeural -f 002.txt --write-media edge.mp3
takes 2 mins to process 30KB text \
Online processing using Microsoft servers

male voices - General and Friendly, Positive unless specified otherwise

en-AU-WilliamMultilingualNeural
en-US-AndrewMultilingualNeural Conversation, Copilot Warm, Confident, Authentic, Honest
en-US-BrianMultilingualNeural Conversation, Copilot Approachable, Casual, Sincere

en-CA-LiamNeural
en-GB-RyanNeural
en-GB-ThomasNeural
en-NZ-MitchellNeural
en-US-AndrewNeural Conversation, Copilot Warm, Confident, Authentic, Honest
en-US-BrianNeural Conversation, Copilot Approachable, Casual, Sincere
en-US-ChristopherNeural News, Novel Reliable, Authority
en-US-EricNeural News, Novel Rational
en-US-GuyNeural News, Novel Passion
en-US-RogerNeural News, Novel Lively
en-US-SteffanNeural News, Novel Rational

female voices - General and Friendly, Positive unless specified otherwise

en-US-AvaMultilingualNeural Conversation, Copilot Expressive, Caring, Pleasant, Friendly
en-US-EmmaMultilingualNeural Conversation, Copilot Cheerful, Clear, Conversational

en-AU-NatashaNeural
en-CA-ClaraNeural
en-GB-LibbyNeural
en-GB-MaisieNeural
en-GB-SoniaNeural
en-NZ-MollyNeural
en-US-AnaNeural Cartoon, Conversation Cute
en-US-AriaNeural News, Novel Positive, Confident
en-US-AvaNeural Conversation, Copilot Expressive, Caring, Pleasant, Friendly
en-US-EmmaNeural Conversation, Copilot Cheerful, Clear, Conversational
en-US-JennyNeural Friendly, Considerate, Comfort
en-US-MichelleNeural News, Novel Friendly, Pleasant