# TalkEasy

Aiolum story: https://aiolum.com/story/talkeasy/ · added 2026-11-05 · updated 2026-11-05

**In short.** TalkEasy (AI English speaking tutor, Canada): US$1.6K a month in recurring revenue, November 2026. Evidence grade A: disclosed by a company or a filing, or verified by a payment processor. Team: 1 person, as listed on TrustMRR in November 2026; the App Store names one developer and seller, Vrishabh Kumar. Launched 26 June 2025; still running, revenue flat.

AI English speaking tutor · Canada · AI-native · solo founder · still running, revenue flat · group: Learning and study

- Outcome: US$1.6K a month in recurring revenue, November 2026 (evidence grade of the headline number: A; headline in US dollars: 18732)
- Launch: 26 June 2025; months from launch to the outcome: 17
- Team: 1 person, as listed on TrustMRR in November 2026; the App Store names one developer and seller, Vrishabh Kumar
- Outside funding: None reported in any source read
- Main factor: Voice practice at a small fraction of a human tutor's price
- Meters, 1 to 4: evidence strength 3, copyable pattern 3, capital needed 1
- Tags: AI-native, Solo founder, Bootstrapped, Running, B2C, Education

Links:

- Visit talk-easy.com: https://www.talk-easy.com/
- App Store: https://apps.apple.com/us/app/id6754468688
- Verified revenue on TrustMRR: https://trustmrr.com/startup/talkeasy

**The product.** An English speaking tutor that runs on a phone or in a browser. The learner talks out loud with one of two animated AI tutors, Sakura or Ken. The app checks each spoken sentence word by word, explains the grammar rule behind each fix, and asks the learner to say the sentence again. A four-minute spoken test places the learner on the CEFR scale before any payment.

**Why it worked.** Many English learners understand more than they can say, and a human conversation tutor costs money and needs a schedule. A voice model can listen and answer at any hour. TalkEasy wraps that in lessons built around the learner's own reasons, a correction view, and a progress page that brings old mistakes back for practice.

Evidence grades: A = disclosed by a public company, a filing or a payment processor, or verified by one; B = two or more independent reports agree, or a journalist confirmed it; C = a founder claim that no one else verified; D = reported, not verified: one founder claim or a third-party estimate with no second source; E = no revenue figure found, so the headline is a non-money fact and the other facts carry their own grades; calc = computed on this page from graded facts.

## Summary

TalkEasy is an AI tutor for spoken English: a program that holds a voice conversation with the learner and corrects the mistakes. It uses a large language model, a program trained on large amounts of text that can write replies; the privacy policy names Google Gemini for the conversations. It also uses speech recognition, which turns the learner's voice into text so each word can be checked. A short spoken test scores grammar, vocabulary, fluency and coherence and places the learner on the CEFR scale, the Common European Framework of Reference, which ranks language skill from A1 (beginner) to C2 (near native).

The App Store lists one developer and seller, Vrishabh Kumar, and TrustMRR lists a team of 1 person, founded on 26 June 2025, in Canada. The site was live by 11 July 2025. Payment records that TrustMRR reads from Stripe and Superwall reported US$1,561 in monthly recurring revenue (MRR, the subscription income that repeats each month) on 5 November 2026, with 159 active subscriptions and US$24,423 in revenue since the start. Monthly revenue fell to US$97.78 in February 2026, then TrustMRR reported between US$1,802 and US$2,415 a month from June to October 2026.

The headline number is grade A, because it comes from the payment processors through TrustMRR. User counts and the founder's description are company statements. The page also shows four failed companies that sold English lessons or language learning, and what is different now.

## Pricing

TalkEasy prices in November 2026, from the US App Store listing and the company's comparison page (grade C)

| Plan | Per month, billed yearly (US$) | Per month, billed monthly (US$) | Limit |
|---|---|---|---|
| Free trial | 0 | 0 | 5 to 7 days, depends on the plan |
| Lite Yearly | 5.00 |  | US$59.99 a year, App Store |
| Annual Special | 8.33 |  | US$99.99 a year, App Store |
| Yearly / Pro Yearly | 10.00 |  | US$119.99 a year, App Store; web yearly shown as US$9.91 a month |
| Monthly |  | 19.99 | App Store |
| Weekly |  |  | US$6.99 a week, App Store |

The App Store lists in-app purchases in US dollars; prices on the web, paid through Stripe, may differ by market, as the TrustMRR profile says. The company's comparison page gives the web yearly plan as US$9.91 a month and a free trial of 5 to 7 days, depending on the plan. Monthly equivalents of yearly plans were computed on this page.

## How much revenue does TalkEasy make? The numbers, graded

A one-person English tutor app with US$1,561 in verified monthly recurring revenue and 159 subscriptions in November 2026, after 17 months

| Fact | Value | Grade | Note | Sources |
|---|---|---|---|---|
| Monthly recurring revenue, 5 November 2026 | US$1,561 | A | TrustMRR metric snapshot from Stripe and Superwall payment records. MRR change over the last 30 days: -7.3%. | [1] |
| Run rate from that MRR | US$18,732 a year | calc | Calculated: US$1,561 times 12. This is the headline number in dollars. | [1] |
| Active subscriptions, 5 November 2026 | 159 | A | TrustMRR metric snapshot. | [1] |
| Average MRR per active subscription | about US$9.82 | calc | Calculated: US$1,561 divided by 159. Close to the yearly plan price divided by 12. | [1] |
| Revenue since the start, to 5 November 2026 | US$24,423 | A | TrustMRR, all-time verified revenue. | [1] |
| Revenue in the last 12 months | US$17,241 | A | TrustMRR. So about US$7,182 came before November 2025 (calculated). | [1] |
| Monthly revenue, December 2025 to October 2026 | US$1,371; 645; 97.78; 750; 1,217; 1,530; 2,096; 2,415; 2,023; 1,802; 2,213 | A | TrustMRR monthly timeline. The best month was July 2026. October 2026 was about 8% below it (calculated). | [1] |
| Revenue in the last 30 days | US$1,922 in the revenue table; US$1,301 in the snapshot line | A | The same TrustMRR profile gives both figures, perhaps taken at different times. The snapshot line shows -19.6% growth. | [1] |
| Registered accounts | about 58,400 | C | TrustMRR profile text written by the founder: 'approximately 58,400 confirmed registered accounts'. The home page says 55,000+ learners and the App Store description says 35,000+ learners. | [1, 3, 2] |
| Share of registered accounts with an active subscription | about 0.27% | calc | Calculated: 159 divided by 58,400. | [1] |
| App Store ratings | 4.6 from 75 ratings (US); 4.8 from 16 ratings (Canada) | C | Read on the App Store pages on 5 November 2026. Age rating 4+. Category Education. | [2, 8] |
| Team | 1 person | C | TrustMRR profile, 'Team size: 1 person'. The App Store names one developer and seller, Vrishabh Kumar. | [1, 2] |
| Founded | 26 June 2025 | C | TrustMRR profile field. An Internet Archive copy shows the site live on 11 July 2025. | [1, 7] |
| AI provider | Google Gemini | C | Privacy policy and terms. Speech recognition comes from other providers that the policy does not name. | [4, 5] |
| Payment providers | Stripe on the web; App Store billing through Superwall on iPhone | A | TrustMRR lists Stripe and Superwall as the verified payment sources. | [1] |
| Legal home | Ontario, Canada | C | The terms are governed by the laws of the Province of Ontario. TrustMRR gives the country as CA. | [5, 1] |
| Time from founding to the headline figure | 17 months | calc | Calculated: June 2025 to November 2026. | [1] |

How to read this table: a grade A value comes from payment records that TrustMRR reads from Stripe and Superwall. A grade C value is a statement by the company on its own pages, its App Store listing or its TrustMRR profile text. A "calc" (X) value was computed on this page from the rows above it.

## When did TalkEasy launch, and how did it grow? Timeline

From a web page in July 2025 to an iPhone app with 159 paying subscriptions in November 2026. Each entry carries its grade.

- **26 June 2025** (grade C [1]): Founding date given on the TrustMRR profile.
- **11 July 2025** (grade C [7]): The site is live: 'Speak confident English in 15 min/day with an AI tutor that answers back', a free first lesson and support for 7 native languages.
- **December 2025** (grade A [1]): US$1,371 in verified revenue, the first month in the public TrustMRR timeline.
- **February 2026** (grade A [1]): Revenue falls to US$97.78 for the month.
- **March to April 2026** (grade C [2]): App updates add type-to-chat, new tutor voices, a free-conversation mode and word-level grammar corrections.
- **May to June 2026** (grade C [2]): The app reaches 8 languages for explanations, then clearer Lite and Pro subscription plans.
- **July 2026** (grade A [1]): Best month so far: US$2,415 in verified revenue.
- **10 October 2026** (grade C [2]): Version 1.10.0: a new design with Home, Explore and Progress pages, and a hands-free mode with no button to hold.
- **5 November 2026** (grade A [1]): US$1,561 in MRR, 159 active subscriptions and US$24,423 since the start, verified through TrustMRR.

## Why did TalkEasy work?

### Speaking, not tapping

Most language apps test the learner with taps and multiple choice. TalkEasy makes the learner talk from the first minute, and the tutor keeps the conversation going. That is the practice a human conversation tutor sells, at a price that the company's comparison page puts at US$9.91 a month on the yearly plan.

### A visible correction for every sentence

Each spoken sentence is checked word by word. The app shows the wrong word, the right word and the rule, then asks the learner to say it again. Fixed mistakes land on a Progress page and come back later as fill-in-the-gap practice, so the learner sees a number go up.

### A placement test before payment

A four-minute spoken assessment gives a CEFR level and scores for grammar, vocabulary, fluency and coherence before the paywall. The learner gets a result first, and the lessons start at that level.

### Short videos and many languages

The founder's profile text says the product is introduced with short videos on Instagram and TikTok. The site is published in 11 languages and the app explains lessons in 8, and the home page says learners come from 100+ countries.

## How it was built

These details come from the TrustMRR profile, the privacy policy, the terms and the App Store release notes. They are company statements, grade C, except the payment providers, which TrustMRR confirms.

- Next.js for the web front end, Python, Supabase for the database and sign-in, and Vercel for hosting, as listed on the TrustMRR profile.
- Google Gemini for the AI conversations, as named in the privacy policy and the terms. Speech recognition comes from third-party providers set to keep no audio and to not train on it.
- Lesson audio is stored so learners can replay it, and deleted after 90 days.
- Stripe takes web payments; Superwall manages the iPhone paywall and App Store subscriptions.
- PostHog for product analytics, OneSignal for push notifications, and TikTok and Meta pixels to measure ads, as listed in the privacy policy's cookie table.
- An iPhone and iPad app, 3D animated tutors with lip-sync, and frequent releases: 25 versions between 13 March and 19 October 2026.

Sources: [1, 4, 5, 2]

## Who tried this before TalkEasy, and why did they fail?

Four companies in our database of failed startups sold English lessons or language learning and failed. The values below are from that database and are not independently verified.

The failed startups come from our database of dead startups, retold here in our own words; Its entries are AI-assisted summaries of public sources, so they carry no evidence grade.

### TutorGroup (iTutor) (2004 to 2021; funding US$300M; cause: Legal/Regulatory)

- Promise: Live online English lessons with native-speaking tutors for learners in Asia.
- Why it died: Every new student needed a paid human tutor, and acquisition costs were high. China's 2021 ban on for-profit tutoring removed most of its market.
- Record: 

### Wall Street English (2000 to 2021; funding US$300M; cause: Legal/Regulatory)

- Promise: Premium English courses for adults in learning centers, paid in advance.
- Why it died: Centers, teachers and sales staff made each student expensive. Regulators stopped the large upfront payments that funded the operation.
- Record: 

### Hujiang (2001 to 2021; funding US$187M; cause: Unit Economics)

- Promise: A large online language-learning community with paid courses.
- Why it died: It reported 190 million registered users, but few of them paid, and the cost of winning each paying customer was higher than what that customer brought in.
- Record: 

### Teacher Finder (2015 to 2019; funding US$50K; cause: Competition)

- Promise: Match language learners directly with tutors, without a language school.
- Why it died: Larger tutor marketplaces offered more, and finding customers became too expensive for a small operation.
- Record: 

## What changed

The failed companies sold time with human teachers, so each new student added a teacher's cost, a schedule and often a building. TutorGroup, Wall Street English and Teacher Finder could not make that cost fall as they grew. Hujiang had the opposite problem: very many free users and too few who paid.

TalkEasy replaces the teacher with a voice model, so a new learner adds only model and hosting costs, and one person can run the business. The Hujiang lesson still applies: TrustMRR reported 159 active subscriptions against about 58,400 registered accounts that the founder listed, so the business depends on turning free users into paying ones.

## Can one person copy TalkEasy now?

**Verdict: Yes. One person built and runs it, and the parts can be bought.**

A voice model, a speech recognition service, a database with sign-in, a payment processor and an iPhone paywall tool are all rented services. The work is in the lessons, the correction view, the placement test and the marketing. The market is crowded, so the copy needs a narrower audience or a better habit loop.

1. Pick one skill that learners pay a human for, here spoken conversation, and let the model do the talking.
2. Show a result before the paywall: a level, a score or a corrected sentence.
3. Keep every mistake and bring it back for practice, so progress is visible.
4. Translate the explanations into the learners' own languages and market with short videos in those languages.

**What is still open.** Speaking practice for one exam (IELTS, TOEFL), one job (nurses, call-center staff, pilots) or one language pair, and the same method for languages other than English.

## What could go wrong for TalkEasy?

- **What is not verified.** The revenue figures are verified by TrustMRR from Stripe and Superwall. The user counts are not: the founder's profile text says about 58,400 registered accounts, the home page says 55,000+ learners, and the App Store description says 35,000+ learners. No interview or article about the founder was found, so the founding story, costs and funding rest on profile fields only.
- **Outcome claims and testimonials without a source.** The App Store description says 'Learners using TalkEasy reach conversational fluency 3x faster. 93% achieve their first learning goal, 89% feel more confident speaking', with no study named, and quotes a testimonial signed only 'David L.'. The home page FAQ says significant fluency improvements typically occur within 4 to 6 weeks. The home page reviews are App Store reviews with usernames and months.
- **Revenue is small and flat.** Monthly revenue rose from US$97.78 in February 2026 to US$2,415 in July 2026, then stayed between US$1,802 and US$2,415. MRR fell 7.3% in the last 30 days, and the same profile gives two different last-30-day revenue figures.
- **The developer's other app.** The App Store page lists one other app by the same developer, 'GAMEFREE: Quit Tracker', described as 'AI therapy & self control'. TalkEasy's own pages do not link to it.
- **Age rating and topics.** The App Store rates the app 4+, and one of its eight topic packs is Dating, practiced as conversation scenarios. The terms read for this page state no minimum age.

## Sources

1. TrustMRR: TalkEasy, public markdown profile, revenue synced 5 November 2026 — https://trustmrr.com/startup/talkeasy.md (grade A). Read the markdown profile. Revenue, MRR, subscriptions and the monthly timeline come from Stripe and Superwall through TrustMRR (grade A); the description, team size, user count, tech stack and marketing channels are founder-written profile text (grade C).
2. App Store (US): Learn English: TalkEasy, by Vrishabh Kumar — https://apps.apple.com/us/app/learn-english-talkeasy/id6754468688 (grade C). Read in full on 5 November 2026. Source for the developer and seller name, the in-app purchase prices, the ratings, the age rating, the version history and the description's claims.
3. TalkEasy home page — https://www.talk-easy.com/ (grade C). Read in full on 5 November 2026. Source for the features, the CEFR assessment, the user counter, the reviews and the FAQ.
4. TalkEasy privacy policy, last updated 22 July 2026 — https://www.talk-easy.com/privacy (grade C). Read in full. Source for Google Gemini, Stripe, the audio retention and the cookie table (Supabase, PostHog, TikTok, Meta, OneSignal).
5. TalkEasy terms of service — https://www.talk-easy.com/terms (grade C). Read in full. Source for Google Gemini, the subscription terms and the Ontario governing law.
6. TalkEasy vs Cambly, the company's comparison page — https://www.talk-easy.com/compare/talkeasy-vs-cambly (grade C). Read in full. Source for the web yearly price of US$9.91 a month and the free trial length.
7. Internet Archive copy of talk-easy.com, 11 July 2025 — https://web.archive.org/web/20250711060624/https://www.talk-easy.com/ (grade C). Read in full. Shows the site live two weeks after the founding date.
8. App Store (Canada): Learn English: TalkEasy — https://apps.apple.com/ca/app/id6754468688 (grade C). Read the ratings and the developer's other apps.
9. Database of failed startups, local export — /method/#dead (grade C). Local export. AI-written summaries, not independently verified. Source for the failed predecessors.

Contact: aiolum@livyden.com. Run TalkEasy? Send corrections or a verified figure to aiolum@livyden.com (subject: Correction: TalkEasy).