Opportunities › Video caption generation and correction › Proposed app
Tallyline
Multi-engine captions with only the doubtful words to check, sized for every screen
- Score
- 71
- confidence 58
- Evidence
- 13
- items on the page
- Rivals
- 3
- named in sources
- First version
- 30 d
- estimate
- Type
- Desktop app
- PWA: runs in the browser, installs on Mac and Windows
- Market since
- 2023
- Emerging market · Verified
ScreensConcept
Desktop app: Video captioning involves uploading files, reviewing dense transcripts, comparing multiple engines' outputs, and exporting in various formats—work best done at a desk with keyboard and larger screen. Sample screens on a computer, in the proposed brand. Their numbers are sample data.
New transcript
Aligning words3 engines
About this appAI proposal
Tallyline turns an audio or video file into accurate captions. It runs your file through several speech-to-text engines, compares them word by word, and asks a language model to settle the words where they disagree, using context and your own vocabulary list.
You review only the flagged words instead of reading the whole transcript. Then you export SRT or VTT files laid out for 16:9, 9:16 or 1:1, with line breaks that keep phrases together and respect limits on characters per line and reading speed.
Built for solo creators, small video teams and production companies who now fix transcript errors and reformat captions by hand. Captions stay as separate track files, so one transcript serves every layout and nothing is burned into the video.
- Tagline
- Captions you only have to check where it matters.
- Voice
- PreciseCalmPractical
- Why this name
- A tally counts agreement between engines, and a line is a caption line. Two plain words, easy to say.
In the first version
Members say which features they would use. It tells PWAS what people want; it changes no score.
Several engines, one transcript
Upload audio or video and transcribe it in parallel with two or three speech-to-text engines, then align them word by word.
No member yet
Disagreements flagged
Words where the engines differ are marked, so you see exactly where errors are likely.
No member yet
LLM settles disputed words
A language model picks the likeliest word from the surrounding sentence and your custom vocabulary of names and terms.
No member yet
Quick review editor
Jump from one low-confidence word to the next, listen to the audio, and fix it in a keystroke.
No member yet
Brand identityAI proposal
Check the name against trademark registers before using it.
Palette
#0F3D3EInk Teal · primary#F7F4ECPaper · surface#1D2426Charcoal · text#E8A317Marigold · accent#D9534FSignal Coral · disputed word flag#7FA99BSage · data
Typography
Aa
Tallyline
PricingAI proposal
The material gives no competitor prices. These tiers, billed by minutes transcribed, are a starting point to test, not a figure drawn from the evidence.
Free
Free
- 30 minutes transcribed
- SRT and VTT export
- One aspect ratio per file
Creator
$15 a month
- 300 minutes transcribed
- All aspect ratios
- Custom vocabulary list
- Flagged-word review editor
Studio
$49 a month
- 1,500 minutes transcribed
- Shared review links
- Team vocabulary lists
- Up to 5 members
Would you pay Creator · $15 a month?
Your answer tells PWAS which prices people accept. It changes no score.
No answers yet.
What it could be worthEstimate
What a software business making this much sells for today, from real deals. Pick a monthly revenue.
If it makes
Worth about
$190,000
Sale range $132,000 to $263,000
$120,000 a year in revenue
From 87 software businesses listed and sold on Empire Flippers in the last year, revenue vetted by the broker: they sell for about 19× their monthly revenue, most between 13× and 26×.
Measured on profit, the middle deal is 32× monthly net profit: about 2.7 years of profit.
What products in this market make
- SubmagicPublic claim$666,667 a month, worth about $12,672,000 at that multiple · getlatka.com
An estimate, not an offer or financial advice. A real price depends on profit, growth, churn, how much of the work is the owner's, and the buyer.
How it would growAI proposal
How the product would reach new customers through its own use, and how the first ones would be found. A plan to test, not a result.
- Built-in loop
- Shared outputs. Exported caption files and an optional review link let editors, clients and collaborators check flagged words in the browser without an account. The shared page carries a small 'Checked with Tallyline' note.
- Do they become customers?
- Reviewers are often clients or editors who could become paying users, but many will only click once. Test how many review-link visitors start their own upload, and whether agencies pass it to other clients.
- First customers
- Solo YouTube and short-form creators and small video agencies (2 to 10 people) who publish both horizontal and vertical versions of the same video.
- Where to reach them
- Creator communities: Offer free trial minutes in video editing subreddits, Discord servers and Hacker News show posts, with before/after error counts.
- Direct outreach to agencies: Message small agencies that post vertical cuts and offer to caption one real project for free in exchange for feedback.
- Comparison content: Publish accuracy tests on public audio, showing flagged words and the fixes, to attract searches on caption correction.
- First milestone
- Ten paying creators or small agencies on a monthly plan, with at least half exporting two or more aspect ratios per video.
Why people already pay for this
- VerifiedHacker News post received 20 points and 2 comments
- AI inferenceVideo creators manually fix both transcription errors and caption formatting
- Public claimSingle transcription models produce errors that require time-consuming correction
- AI inferenceExisting tools do not handle caption formatting well across different screen shapes
- Public claim
Tech stack
Every proposal builds on the same PWA base, installable from the browser on any phone or computer. Dashed boxes are what this app adds.
- AppNext.js and TypeScript
- PWAManifest and service worker
- UITailwind CSS
- Data and authSupabase Postgres
- PaymentsStripe
- EmailResend
- HostingVercel
- This app · Speech-to-textTwo or three third-party transcription APIsCombining independent engines is the core of the accuracy claim.
- This app · Language modelHosted LLM API for word arbitrationSettles disputed words using context and the vocabulary list.
RisksAI inference
- CommercialWell-funded caption apps with styled, animated captions already attract many creators. Tallyline skips styling in version one and must win on accuracy and review time alone.
- TechnicalAligning word-level output from different engines, with different timestamps and spellings, is hard. Poor alignment would flag too many words or miss errors.
- CommercialRunning several engines plus an LLM per file costs more than a single engine. Per-minute margins at these prices are untested.
- AIThe LLM may confidently choose a wrong word or rewrite what was actually said. Show the original engine candidates and keep every change reviewable.
- PlatformThe product depends on third-party speech-to-text and LLM providers, whose prices, terms and quality can change.
Everything marked AI proposal or AI inference was written by a model from the evidence on the opportunity page. The proposed name must be checked against trademark registers before it is used.