Frequently asked questions

Anyone aged 18 or above who can speak at least one of our supported dialects fluently, has a smartphone with internet, and has a UPI ID in their own name to receive payments.

Recording on LootBeez is by invitation only, so that every dataset we deliver comes from real, verified native speakers. Anyone can sign up and complete their profile, consent and training, but you can start recording only after our team approves your account. You will get an email and an in-app notification as soon as you are approved. Approval is free — we never charge any fee, and no one from LootBeez will ask you for money to approve your account.

Yes. There is no joining fee, deposit, training fee or subscription — not now, not later. We earn when our business customers pay for data, and you earn your share for each approved recording. Anyone asking you for money in the name of LootBeez is a fraud.

Enter your email address and we send you a 6-digit code. There is no password to remember. Codes expire after 10 minutes. Never share the code with anyone.

Speech technology must work for everyone, so customers need recordings from a balanced mix of speakers. We share only your age band (for example 25–34), gender, state, district and dialect — never your exact age, name or contact details.

Before your first task you answer six short questions about how to record well and our basic rules. It takes about two minutes. You can retake it as many times as you need.

No. You can use LootBeez in your phone's browser at lootbeez.shop/app and add it to your home screen. An Android app is also available.

Reading sentences aloud, speaking voice commands, describing a picture, talking about a topic for a minute, and listening to recordings and writing what is said (transcription). Each project shows its task type and instructions.

• Sit in a quiet room — turn off the fan and TV.
• Hold the phone about 15–20 cm (one hand span) from your mouth.
• Read exactly what is written, in your natural accent.
• If you make a mistake, tap "Redo".
• Take a short break every 20–30 minutes.

No. Each person must create their own account and record only their own voice. Recordings from other people on your account will be rejected and may lead to account suspension. Every adult in your family is welcome to join with their own account.

Yes. Each project has a limit per contributor so that the data includes many different voices. When you reach it, you will see a message and can choose another project.

Your recording is saved on your phone and uploaded automatically when you are back online. You will see a banner showing how many recordings are waiting.

Common reasons are background noise, low volume, reading a different sentence, a recording that is cut off, speaking a different dialect, or a duplicate. The exact reason is shown in "My recordings". Rejected recordings are not paid.

After at least 20 of your items have been reviewed, your approval rate over your last 200 reviewed items sets your level. Silver (80%+) and Gold (92%+) earn an extra bonus on every approved recording — currently 5% and 10%. Bronze earns the base rate. The current thresholds are shown in the app.

Trained reviewers who speak the same language. Reviewers can never review their own work. Some reviewers are experienced contributors who were invited to the reviewer role.

Write to us from Help in the app within 15 days, mentioning the recording. We will re-check it. If the rejection was a mistake, we will approve it and pay you.

You earn the project's rate for the duration of your approved audio. For example, at ₹360 per approved hour, 10 approved minutes earn ₹60. Rates differ by project and are shown before you start. We do not guarantee any minimum earnings.

Once a week, on the payout day shown in the app, we pay all balances above the minimum payout to your UPI ID. Smaller balances carry forward to the next week.

You receive an in-app notification and an email with the amount and UPI reference (UTR). Every payout is also listed under Earnings → Payouts.

Indian tax law may require us to deduct tax at source (TDS) and report payments above certain limits. When your total earnings cross the limit shown in the app, we need your PAN to continue paying you. Your PAN is never shared with customers.

The amount returns to your balance and is retried in the next payout. Please check that your UPI ID in Profile is correct and active.

Share your code or link. When a friend who joined with it gets their first 30 minutes of recordings approved (current setting), you receive the referral bonus. Self-referrals and fake accounts are not allowed.

Companies, data vendors and research institutions building speech and language technology. They receive the audio and pseudonymous metadata under a licence that forbids trying to identify you. See the Contributor Agreement.

Yes. Go to Profile → "Delete my account & data". Recordings that have not been delivered to a customer are deleted. Recordings already delivered cannot be recalled, but they carry no personal identifiers.

No advertising or tracking cookies. We store your login session on your own device. See our Cookie & Storage Policy.

16 kHz mono WAV files with CSV/JSON metadata (pseudonymous speaker ID, age band, gender, state, district, dialect, prompt and/or transcript), delivered in milestones with a QC report.

Every contributor accepts a versioned consent and rights assignment before recording. Records are time-stamped and retained, and can be summarised in your delivery report.

Yes. Each project comes with a private, read-only dashboard link showing approved hours, speaker demographics and daily progress.

Usually an advance on order and the balance on milestone delivery, as stated in your order. See our Data Licence Terms and Refund & Cancellation Policy.