Capflair
PricingChangelogAbout
Sign inStart free
LEGAL

Privacy Policy

Last updated, 27 October 2026

Account information

  • When you create an account with email and password, we store your email address and a securely hashed password.
  • When you choose Continue with Google, Google shares your name, email address and profile picture with Capflair. We request only these basic profile scopes (email, profile), and nothing else from your Google account.
  • Accounts and sign-in sessions are managed by our authentication provider, Supabase, on servers in the Mumbai (ap-south-1) region.

How we use Google user data

  • Google account data is used only to create your Capflair account, sign you in and show who is signed in.
  • We do not sell Google user data, use it for advertising, or share it with anyone except the service providers needed to run sign-in.
  • Capflair’s use of information received from Google APIs adheres to the Google API Services User Data Policy, including the Limited Use requirements.

Your video and audio

  • Your video stays in your browser. Previews and exports are rendered on your own device, and the video file is not uploaded to Capflair.
  • To create captions, the audio track is extracted in your browser and sent in short pieces to our transcription providers: Deepgram, and ElevenLabs for song and lyric passages. They return the words and timings; Capflair does not keep the audio.
  • If you use "Fix spellings with AI", the caption text (in Urdu and Roman Urdu) is sent to OpenAI to correct its spelling. If you use the hook generator, the opening lines of your transcript are sent to OpenAI to suggest hooks.
  • Captions, styles and saved looks are kept in your browser’s local storage on your device.

Plan and usage

  • We keep your plan and this month’s usage (seconds of audio transcribed, seconds of song passages, and number of AI requests) with your account, to apply your plan’s monthly limits. Audio is not stored with these counters. To avoid repeat AI charges, we also keep account-specific request hashes and AI text results. Results can be reused for 30 days; expired cache records are removed when the account next uses AI. Failed requests can be reused for five minutes. New AI attempts count towards the allowance, including attempts that fail after being reserved; cached repeats do not.
  • For essential service operation and cost monitoring, provider request records include your account, plan, provider and model, feature, request status and duration, audio duration or token counts where available, and cache or limit outcomes. They exclude audio, caption text, prompts, file names and API keys. We retain this operational telemetry for approximately 13 months, removing older records periodically as the service is used. It is separate from optional visitor analytics. Cost figures are estimates, not provider invoices.

Error reports

  • When something fails, such as a transcription or an export, the app sends a short technical report so we can fix it: the error message, the page, your browser type, basic settings like file size and export quality, and your account ID if you are signed in.
  • Error reports never include your video, audio or caption text. They are kept for up to 90 days.

Service providers

  • Supabase (accounts and sign-in), Vercel (website hosting), Google (Sign in with Google), Creem (card checkout, billing and tax), Deepgram and ElevenLabs (speech-to-text), and OpenAI (optional spelling fixes and hooks). Each processes data only to provide its part of the service.

Cookies and analytics

  • We use essential cookies to keep you signed in. They are required for your account to work.
  • We count page visits with Vercel Web Analytics, which uses no cookies and does not identify you: the page address (without project ids or other query details), the site that sent you, campaign tags in the link, your country, device type and browser. When you create an account we also save, once, the site and campaign tags of your first visit, so we know which channels bring new users.
  • We also record which product steps happen, so we can see where people get stuck: pages opened, sample actions, uploading a video, generating captions, export attempts and outcomes, requesting downloads, editing captions, using AI tools, picking a style or font and opening checkout. Each record holds the step, the page address, the site and campaign tags that sent you, your country, device type, browser and operating-system family, small details such as video length, operation duration, error category or export quality, an anonymous id stored in your browser, and your account if you are signed in. Custom font names, video, audio, caption text and file names are never included. Records are retained for approximately 13 months and older records are removed periodically as the service is used.
  • Editor step details also include audio readiness and a codec category, whether matching audio was attached, the current editor step, export stage timing, no-progress warnings and cancellation or retry outcomes. The audio check runs on your device. These records do not include file names, caption text, uploaded font data or logo images.
  • Choose No thanks in the consent banner to turn both off on that browser. Declining does not affect the caption editor.
  • When configured, Microsoft Clarity is an additional, optional tool for masked session replays and heatmaps on public pages. It loads only after you separately opt in to replays; an earlier OK for visit counts does not enable it. This first rollout excludes the sample editor, private editor, projects, sign-in, invite, account and payment pages, as well as addresses with query parameters or fragments. Existing product-step analytics continue to measure editor outcomes under the choices described above.
  • Clarity receives masked public-page layout and interactions, page addresses, referring page, device and browser information, and uses cookies after replay opt-in. Advertising storage is denied. We do not send account identifiers or event properties to Clarity. Capture ends before a page transition and checks the new page before restarting. Pages reached from a private address or a referring address containing query details are also excluded. You can withdraw replay consent from Analytics preferences on this Privacy page, or turn all optional analytics off with No thanks. Withdrawing stops future capture; it does not delete already collected recordings. Clarity normally keeps recordings for 30 days, with some recordings retained for up to 9 months.

Payments

  • Card checkout is handled by Creem, our Merchant of Record. We send your account email and an internal purchase reference to create a checkout. Creem collects card and billing details; Capflair never receives or stores your full card number. We keep the plan, amount, currency, payment reference and confirmation status to activate access and handle refunds or disputes.
  • For local transfers we accepted earlier, we keep the transaction ID, the payment screenshot you submit, and any sender details you provide so we can verify the payment. Screenshots are private and available only to you and authorized payment reviewers.
  • We use your country, as provided by our hosting service, to show a local currency estimate next to USD card prices. Exchange rates are fetched by our server from ExchangeRate-API without sending your account details or IP address to that provider. Your bank determines the final currency conversion and fees.

Your choices and deletion

  • You can sign out at any time from the editor.
  • To delete your account and its data, email us from the address on your account and we will delete it. You can also remove Capflair’s access from your Google account at myaccount.google.com/permissions.
  • Clearing your browser’s site data removes the captions and styles stored on your device.

Changes to this policy

  • If we change how we use your data, we will update this page and the date above before the change takes effect.

Questions? support@capflair.com

© 2026 Capflair.PrivacyTerms