Learning and careers decision

Turkishfluent

A competent developer can reproduce a useful subset (web lessons, audio playback, STT-based pronunciation feedback) in multi-week effort using open-source building blocks, but the full paid product (polished mobile apps, curated content and human support) is not practical to match alone.

Visit website

Built by Nicolas Le Roux 🇪🇺, who ships 5 products in this index

You pay

$11.9/mo

$143/yr

Read off the official pricing page.

You’d pay instead

$100one-off82 h to build

$15/mo6 h/mo upkeep

On cash alone, building overtakes the subscription at 2 seats.

Open-source builds that already do this

Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need - the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship. All Turkishfluent alternatives, with the arithmetic →

What a replacement has to do

  • Deliver a daily 10-minute lesson (dialogue audio + transcript), let the user play/record audio for speaking exercises, send the recording to a speech-to-text/pronunciation scoring service, store progress and show simple stats, unlock next lesson.

What it still won’t have

  • Polish and UX of mobile native apps (App Store / Play Store experience)
  • Curated, pedagogically sequenced content and editorial effort
  • Human teacher / support included in paid plan
  • Production-grade reliability, scaling, and analytics dashboards
  • Brand, marketing reach, and user trust

What remains hard

  • Product polish and ongoing maintenance
Read the build prompt

First-year cost

Keep paying

Paying is—cheaper in year one.

On cash alone, building overtakes the subscription at 2 seats.

Paid seatsseats

Money you would actually spend

Keep paying
—

Subscription price × seats × 12

Build it
—

AI build —APIs + hosting —

Time you would spend

—

—

What you would spend

What we assumed

The verdict above measures whether you could build it. This one is only about money.

Runnable build prompt

Not run yet
Build a minimal TurkishFluent replacement as a web app using Next.js + React, PostgreSQL (Supabase optional) and Stripe for billing. Scope: (1) admin CMS to upload lessons (title, transcript, audio files), (2) public responsive lesson pages with audio playback and dual-speed playback, (3) user auth, 3-day trial, and Stripe monthly subscription checkout, (4) record-or-upload audio in-browser, send recordings to OpenAI Whisper (or Whisper-compatible) for STT and compute simple pronunciation feedback (compare phoneme-level confidence or word-error and return highlighted transcript), (5) store user progress and simple per-lesson stats, (6) include automated tests for auth, billing flow, lesson CRUD, and the STT scoring pipeline, and basic error handling (upload failures, API timeouts). Out of scope: native mobile apps, advanced adaptive learning algorithms, paid human tutoring. Deliver docker-compose for local dev and deployment instructions to Vercel (frontend) and a small VPS or Supabase (backend), include a README with cost estimates for hosting and APIs.
How we checked4 sources · 2/3 runs agreed · evidence score 63

How the score was reached

  • Partly verdict base52
  • An open-source build was found+5
  • 4 cited sources+3
  • Price verified on pricing page+3
  • Evidence score63

The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time - so the same evidence always produces the same number.

How scoring works →

Cited sources · 4

Every page the run actually retrieved.

Integrity checks

What held up, and what did not.

✓ Price read off the page! 2 of 3 runs agreed; the verdict is the majority✓ Citations limited to fetched pages! 1 moat recorded