Your questions, answered.

Fluento AI is a voice-first English speaking app built for Indian learners. You talk to AI characters like Maya, Rajesh, and Emily in real scenarios — interviews, work calls, café orders — and get instant CEFR-aligned feedback on your fluency and accuracy. It works on Android and iOS and starts free.

Install Fluento AI from the Play Store or App Store, allow mic permission, and pick any scenario. The AI character speaks first — you reply, it corrects gently, and the conversation keeps going. Most users finish their first session in under 5 minutes.

Yes. You get 15 minutes of voice practice and 3 detailed corrections every single day — forever. No credit card, no trial clock, no ads. Upgrade to Pro only when you want unlimited minutes and advanced scenarios.

Any level from A1 beginner to C1 advanced. The AI adapts to you. If you can read this sentence, Fluento AI can teach you. Your CEFR level updates after your first 3 conversations.

15 to 20 minutes. Research on language acquisition shows that short daily sessions beat long weekly ones. Fluento AI's free plan is designed around this — 15 minutes a day is enough to jump one CEFR sub-level every 4–6 weeks.

Fluento AI has three core personas with distinct accents and personalities: Maya (27, Mumbai journalist, warm), Rajesh (42, Bangalore HR manager, professional), and Emily (29, London designer, direct). More characters ship every month, including US, Singaporean, and Delhi voices.

80+ and growing. From airport check-in at Delhi to salary negotiation, Infosys HR round, first-date coffee, US B1/B2 visa interview, and haggling at Chandni Chowk. Pro unlocks the full library; free tier rotates 8 scenarios each week.

CEFR (Common European Framework of Reference) is the global standard for language proficiency: A1, A2, B1, B2, C1, C2. Fluento AI scores every turn you speak on vocabulary, grammar, fluency, and pronunciation, then maps a weighted average to CEFR. Your level re-evaluates after every session.

Yes. Fluento AI uses frontier LLMs trained on ESL error data, cross-checked against CEFR descriptors. We audit 0.5% of sessions weekly against human raters and target 94%+ agreement. If you disagree with a correction, tap the thumbs-down — it trains the model for your account.

Yes. When you hit B1, B2, or C1 consistently (across 10+ sessions), Fluento AI generates a CEFR certificate with a unique verification URL. Indian employers — especially in BPO, airlines, and international roles — recognise CEFR.

Indian, British, and American. Each character has a distinct native accent so you train both production and perception. Pro users can lock any scenario to a specific accent — useful if you are preparing for a UK visa or a US tech interview.

Yes. Rajesh's 'Mock Interview' scenario simulates a full 20-minute HR round with real Infosys, TCS, and Wipro-style questions. You get a score on clarity, structure (STAR), and confidence, plus a replay of your weakest answer.

Your dashboard shows CEFR level, weekly minutes, streak, and a ribbon-chart of the top 5 skills you are improving — e.g. past-tense accuracy, filler-word reduction, speaking rate. Every Sunday you get a progress summary on WhatsApp if you opt in.

Pro is ₹199/month. The Annual plan is ₹1,499/year (about ₹4/day). A one-time Lifetime plan is ₹4,999. All prices are inclusive of GST. Payment is via UPI, cards, or Razorpay wallets.

Unlimited voice minutes, all 80+ scenarios, advanced corrections (native-level nuance), certificate generation, accent-lock, interview deep-dive mode, priority mic latency, and early access to new characters.

Yes. Cancel from the app in two taps. Your Pro features stay active till the end of the billing period. We do not ask why, we do not send guilt emails.

Full refund within 7 days of purchase, no questions asked. For annual and lifetime plans, partial pro-rata refunds within 30 days. Email support@fluentoai.com or WhatsApp us and the refund lands in 3–5 working days.

Yes. College students with a valid .edu.in or .ac.in email get 40% off annual — ₹899 instead of ₹1,499. Write to support@fluentoai.com from your student email.

Family (up to 5 members) is ₹2,499/year. For colleges and corporate teams of 20+, our Teams plan includes a learning dashboard, custom scenarios, and LMS integration. Contact us from the /for-teams section.

About 15 MB per 10-minute session — similar to a WhatsApp voice call. You can enable 'Lite audio' in settings to cut this to 8 MB. Fluento AI works well even on Jio 4G in Tier-2 cities.

Partial. You can revisit past conversations and review corrections offline. Live voice sessions need an internet connection because the AI model runs on our servers. We are testing an on-device mode for 2026.

A 20-minute session uses roughly 3–4% battery on a mid-range Android (Poco, Redmi, Realme). We throttle background animations aggressively on devices below 4 GB RAM.

To hear you. Audio is streamed to our servers only while a session is active, and is deleted after the transcript is generated unless you explicitly save it for replay. We never access the mic in the background.

Android 8+ and iOS 14+. Works on phones as old as Redmi Note 7 and iPhone 6s. Tablet support is live on both platforms. A web app is on the 2026 roadmap.

Yes, directly. Fluento AI detects common MTI patterns — v/w swaps (Hindi), 'simple past as present' (Tamil, Telugu), 'itself/only' filler (pan-Indian) — and coaches you with drills tuned to your L1. You can set your first language in settings.

No. Indian English is a legitimate dialect. Fluento AI marks pronunciation only when it affects intelligibility to a non-Indian listener — e.g. if 'development' sounds like 'develop-mint'. Accent ≠ error.

Yes — this is our strongest use case. The Mock Interview scenario uses Infosys, TCS, Wipro, Accenture, and Cognizant-style questions. You get STAR-method coaching, filler-word counts, and a confidence score after every round.

Yes. There are dedicated scenarios for US B1/B2, US F1 (student), UK Tier 2, and Canada study-permit interviews. You rehearse with Rajesh playing a skeptical consular officer. Not a guarantee of visa success, but strong practice.

The UI is available in English, Hindi, Tamil, Telugu, Bengali, and Marathi. The AI always speaks in English (that's the point), but it can explain corrections in your chosen UI language.

Yes. 60% of our beta users are from Tier-2/3 towns — Nashik, Coimbatore, Bhubaneswar, Kanpur, Indore. Lite audio mode and offline review mean you do not need constant high-speed internet.

Transcripts are stored so you can review past sessions. Raw audio is deleted within 24 hours unless you tap 'Save replay'. You can bulk-delete everything from Settings → Privacy → Delete my data — live endpoint, instant action.

No, not by default. Your audio is not used for model training unless you explicitly opt in via Settings → Contribute to research. Fewer than 3% of our users opt in.

Yes. We have a named Grievance Officer, explicit consent flows, data-localisation options, and breach-notification protocols — all documented in our Privacy Policy. Minors (under 18) require parental consent by default.

The app is rated 12+. Under-18 accounts require parental consent in India (DPDP) and COPPA-style consent elsewhere. Child accounts have no social features, and all scenarios are age-appropriate.

Settings → Privacy → Delete my data, or visit fluentoai.com/privacy/delete-my-data from any browser. Your account and all associated audio/transcripts are erased within 30 days (24 hours for audio).

Duolingo teaches English through gamified vocabulary quizzes — great for streaks, weak for speaking. Fluento AI is voice-first: every session is a real spoken conversation with CEFR feedback. See our full comparison at /vs/duolingo.

Cambly connects you to a live human tutor for about ₹800–1,500/hour. Fluento AI gives you unlimited AI practice at ₹199/month. No scheduling, no small-talk tax, no judgement — and one year of Fluento AI costs less than one Cambly session. See /vs/cambly.

ELSA focuses on pronunciation drills (phoneme-level). Fluento AI focuses on full conversations — useful when you can pronounce words but freeze in a real interview. We cover pronunciation too, but the unit of practice is a dialogue. See /vs/elsa.

YouTube is passive. You cannot practise speaking by watching someone else speak. Fluento AI forces you to produce English and corrects you in real time — the only way to build fluency. Use YouTube for theory, Fluento for practice.

Different. A good human tutor is irreplaceable once a week. But humans are expensive, scheduling is a pain, and most of us hesitate to make mistakes in front of a person. Fluento AI is the low-friction daily reps — use a human tutor for feedback once a month if you can afford one.