FindTube
FindTube is a semantic search engine for educational video: it searches the spoken subtitles of a curated library of lectures and podcasts, then jumps the player to the exact timestamp. Built for students, researchers and lifelong learners.
What is FindTube?
FindTube is a browser-based search engine for educational video, published by Future Study Space Pte.Ltd., a Singapore company. Its central idea is simple: rather than matching your query against titles, tags and uploader descriptions, it searches what is actually said, subtitle line by subtitle line, across a curated library of lectures and podcasts. The publisher describes the engine as an X-ray of educational content rather than a directory of it. What comes back is not a list of videos to sift through. Open a result and the player loads with the progress bar already positioned at the moment the answer begins. FindTube is explicit that it does no cutting: it never isolates or downloads the underlying video file, so the original material stays whole and the academic context around a segment is preserved. It is a navigator, not an editor. The library is closed and curated — educational videos and podcasts only, entertainment deliberately excluded. Subjects sit under STEM & Tech (mathematics, physics, chemistry, astronomy, environmental science, engineering, computer science, artificial intelligence, robotics, blockchain and crypto), Economics, Business, Health & Life Sciences (biology, biotechnology, medicine, neuroscience) and Humanities & Social Sciences (psychology, sociology, anthropology, political science, law, linguistics, philosophy, history, literature, film analysis, art history, music theory), plus Trending and Others. Results can be narrowed by school level — All, Primary, Middle, University, Graduate — and by period. Once you are on the playback page a second layer takes over, a study toolkit the publisher calls an academic copilot. It generates a structured viewing plan citing several videos, surfaces counter-arguments and opposing views, maps the prerequisites and next steps around a topic, suggests semantically related videos chosen on meaning rather than popularity, produces mind maps, bullet summaries, curated study plans and critical-thinking questions, and exports the full transcript. The technical claims behind it are contextual NLP extraction, dynamic semantic segmentation, millisecond media synchronisation and handling of specialised scientific terminology. Access is by browser only — no API, no mobile app, no extension — and both the interface and the content are in English. The terms state plainly that matching is done by non-human retrieval methods and that generated content comes from a probability model, with no guarantee of accuracy.
What it does
- Search a curated educational video library in plain language, by question or by concept
- Jump straight to the timestamp where the answer starts, instead of watching the whole video
- Filter by discipline, school level (Primary, Middle, University, Graduate) and time period
- Generate a structured viewing plan that cites and links several videos on one topic
- Surface counter-arguments and opposing viewpoints on the subject you are studying
- Produce mind maps and bullet summaries of the segment you are watching
- Export the full transcript to DOCX, Markdown or PDF
When to use FindTube / When not to
A quick filter to help you decide if FindTube is the right fit.
When to use FindTube
- University and graduate students hunting for one specific concept inside a two-hour lecture, without watching it end to end
- Exam candidates revising an academic subject who want to move quickly between several explanations of the same idea
- Researchers running a video literature review and looking for opposing views on a claim
- Self-taught learners who want a structured path through a topic, from prerequisites through to the next steps
- Note-takers and writers who need full transcripts exported to DOCX, Markdown or PDF for offline work
When not to use FindTube
- Anyone after entertainment, news or general-interest video: the library is deliberately restricted to educational content
- Video editors and creators who need to cut, isolate or download clips — FindTube navigates videos, it never produces a file
- Non-English speakers: the interface and the indexed content are English only
- Developers and teams wanting an API, a mobile app or a browser extension, none of which exist
- Organisations buying under compliance or commercial constraints: there is no data processing agreement, no named subprocessor list and no stated hosting country, and the terms forbid commercial reuse of generated output without written permission
How to use FindTube
A typical end-to-end flow, from setup to results.
- Open FindTube in any browser — there is nothing to install and nothing to configure
- Type your question or concept in plain English; the free tier lets you search without an account
- Narrow the field with the discipline menu, the school-level filter (All, Primary, Middle, University, Graduate) and the time-period filter
- Or browse by discipline from the menu, or start from the Trending page when you have no precise question
- Let the engine scan the spoken subtitle text rather than titles, tags and descriptions
- Open a result: the player loads and the progress bar jumps to the segment where the answer starts
- Work the segment with the study tools — viewing plan, counter-arguments, prerequisites and next steps, related videos
- Generate a mind map, a bullet summary or critical-thinking questions from the subtitles
- Export the full transcript to DOCX, Markdown or PDF when you need it offline
- Create an account with Google, Apple or an email and code when you want to subscribe; cancel later from Settings, under Instructions for Account Cancellation
Pros & Cons
Pros
- Search runs on what is actually said, not on titles, tags or uploader descriptions, so you do not have to guess the right keywords
- Real time savings: you land on the formula or the definition instead of scrubbing through a two-hour lecture
- The original video stays whole and playable, so the academic context around a segment is preserved
- The library is curated and excludes entertainment, which keeps results far cleaner than on a general-purpose platform
- Counter-argument recommendations are a deliberate push against the echo chamber
- A permanent free tier covers basic search and the AI matching features, with no commitment
- Transcripts export to DOCX, Markdown and PDF, and the discipline tree and school-level filters are unusually granular
Cons
- No pricing page and no amount published anywhere on the site: you cannot know what a subscription costs before committing to it
- No postal address for the publisher, no social or community channel, and a single email address serving support, legal and privacy alike
- The GDPR is never mentioned, and no data protection officer or Article 27 representative is designated
- No data processing agreement, no named subprocessors and no stated hosting country
- Browser only: no API, no mobile app and no browser extension
- English only, and the size of the library is never quantified, so coverage of a given subject is hard to predict
- Subscriptions renew automatically and are non-refundable outside two narrow cases, and the terms reserve the right to change or discontinue the service at any time, with or without notice
Pricing & Plans
FindTube operates on a freemium basis. A permanent free tier is available and covers basic search together with the AI matching features; the terms present it as a free credit whose amount and period of validity the publisher sets unilaterally and displays on the relevant page. Once that credit is used up, continued use requires a paid subscription. No price is published: the site carries no pricing page and states no amount, no currency and no billing period anywhere, so no entry price can be quoted. Subscriptions renew automatically for periods of equal length, monthly in the absence of any other indication, and are non-refundable except in two cases: a duplicate charge, or an interruption of service exceeding ten consecutive days and attributable to the publisher. Taxes are payable by the user, payment is handled by an unnamed third-party processor, and access stops immediately when the subscription expires or the balance runs out.
- Free tier — basic search and the AI matching features
- usable without a subscription
- presented in the terms as a free credit whose size and duration the publisher sets
- Paid subscription — continued access once the free credit is used up
- the publisher gives it no commercial name
- details no scope and publishes no amount
- currency or billing period
Data, GDPR & hosting
A consolidated view of how FindTube handles your data.
GDPR overview
FindTube makes no GDPR claim of any kind. The privacy policy and the terms, both effective 29 January 2026, never name the GDPR or any other data protection regulation, designate no data protection officer and no Article 27 representative in the EU, state no legal basis for processing, and say nothing about international transfers or standard contractual clauses. Governing law is Singapore. The policy does, however, describe deletion and anonymisation rights on grounds that mirror GDPR wording: data no longer necessary, consent withdrawn with no other legal basis, objection without overriding legitimate grounds, and unlawful processing. No response deadline is promised and no dedicated request form exists; the only channel is the support@findtube.ai address. Read the alignment as informal rather than contractual.
Who owns the data?
Ownership splits three ways under the terms and the privacy policy, both effective 29 January 2026. Third-party documents surfaced by the search remain the property of their original right holders, and Future Study Space Pte.Ltd. expressly claims no intellectual property in them. Content produced by the service itself — answers, slide decks, spoken explanations — is claimed by the publisher, which grants users the right to read, save and share it for non-commercial purposes only. Material you supply (account details, payment data, uploaded documents and images, reviews and comments, microphone input, the content of messages you send) is collected and processed by the publisher; you may withhold it, at the cost of the features that depend on it.
Reuse rights
Reuse by the end user is permitted but narrow. Output generated by FindTube may be read, saved and shared for non-commercial purposes without asking anyone; anything commercial requires the publisher's written permission, and systematically harvesting results to build a collection, compilation, database or directory is prohibited outright. Videos and documents found through the search stay under the rights of their original owners, so republishing them is not FindTube's to authorise. On the publisher's side, personal data is used to identify and authenticate users, run the service and process orders, communicate, improve the product, recommend content, enforce the publisher's rights against fraud, spam and abuse, and meet legal obligations. It may be shared with service providers, third-party vendors and business partners, with affiliated companies, with the user's consent, in a business transfer, and with law enforcement — the policy undertaking to notify the user promptly and forward a copy of any law-enforcement request unless legally prohibited. None of those providers is named.
Data retention & training
Hosting summary
FindTube says nothing about where its data is hosted. No hosting country or region is named anywhere on the site, there is no trust or security page, and no certification is displayed — no SOC 2, no ISO. The privacy policy, effective 29 January 2026, refers to service providers, third-party vendors and business partners, and to a third-party payment processor, but names none of them and publishes no subprocessor list. It also mentions backup archives in which personal data may survive a deletion request: where deletion is not possible, the data is secured and isolated from further processing until it can be removed. Security is described only in general terms — reasonable and appropriate steps against unauthorised access, use, disclosure, alteration and destruction — with users advised to choose a strong, unique password. Governing law is Singapore and the publisher, Future Study Space Pte.Ltd., is a Singapore company, but that is a legal indication, not a technical one: it says nothing about where the servers are. Anyone with a data residency requirement should ask before signing up.
Things to keep in mind
Risks and trade-offs to weigh before adopting FindTube.
- Generated output comes from a probability model and the publisher guarantees no accuracy: verify anything you intend to rely on
- Subscriptions renew automatically for periods of equal length and are non-refundable outside a duplicate charge or an outage of more than ten consecutive days attributable to the publisher — cancel before the renewal date
- The publisher may modify or discontinue the service at any time, with or without notice, and its aggregate liability is capped at what you paid in the twelve months before a claim
- Commercial use of generated output requires written permission, and building a derivative collection, compilation, database or directory from results is prohibited
- Automatic collection is broad: IP address, device identifiers including MAC address, browsing and click history, search keywords and documents consulted
- No hosting country is stated and no subprocessor is named, so you cannot know where your data sits or who handles it
- Landing straight on the answer is efficient, but leaning on it habitually can crowd out the reasoning that surrounds it; the counter-argument tool exists precisely because a search that only ever confirms you is a comfortable trap
Setup & Integrations
Technical difficulty
Very low. FindTube runs entirely in the browser: nothing to install, nothing to configure, no API key and no integration to set up. An account is not required for basic search, and creating one takes a single step through Google or Apple, or an email address with a code. Queries are written in plain English, so there is no search syntax to learn. The only permission the product asks for is microphone access for voice input; refusing it disables that one function and leaves the rest of the product intact.
Deployment
Integrations
Supported languages
Behind FindTube
Resources
All the official URLs gathered for verification and reference.
Alternatives
Tools that compete with or complement FindTube.
Frequently asked questions
What exactly does FindTube search?
Do I get a video back, or a moment?
What kind of content is in the library?
Is FindTube free?
How much does the subscription cost?
Is there an API or a mobile app?
What does it do beyond search?
Who publishes FindTube, and who may use it?
Does FindTube comply with the GDPR?
Should you pick FindTube?
FindTube does one thing and defines it clearly: it searches the spoken text of a curated educational library and drops you at the right moment. The premise holds up. Searching subtitles instead of metadata removes the keyword guesswork that makes general video platforms frustrating for study, and the refusal to cut or download anything keeps the original lecture intact around whatever segment you land on. The post-playback toolkit — viewing plans, prerequisite maps, counter-arguments, mind maps, transcript export — pushes the product past search and into study, which is where its real value sits. The reservations are about the company, not the engine. The service is young: the domain was registered on 23 January 2026, the terms and privacy policy took effect on 29 January 2026, and the site has only been publicly visible since 21 February 2026. It is commercially opaque, with no pricing page, no published amount and no named plan, so nothing can be compared before signing up. It is institutionally opaque too: no postal address, one email address for support, legal and privacy alike, and no social or community presence. And the compliance layer is thin — no GDPR mention, no data processing agreement, no named subprocessors, no stated hosting country. The publisher is at least candid about the technology, stating that matching is done by non-human retrieval and that generated content comes from a probability model with no accuracy guarantee, which is the right way to read every summary and study plan it hands you. Worth trying on the free tier, particularly if you study from long-form video. Worth more caution before paying, and not a serious procurement candidate for any organisation that needs contractual guarantees about its data.
- Choosing a selection results in a full page refresh.
- Opens in a new window.