Getting started
What exactly is transcriptfy?
transcriptfy is an automatic transcription service that converts your audio and video files to text using artificial intelligence. You upload a file, we process it in seconds, and we give you back the text with timestamps, speaker identification, and export options in multiple formats.
It's designed for journalists, podcasters, researchers, lawyers, students, and anyone who spends too much time manually writing down what someone said.
Do I need to create an account to try it?
No. You can transcribe a sample of up to 30 seconds without signing up — that's the guest mode. If you like the result and want to process the full file or access the editor, we'll ask you to create an account at that point. When you sign up, the transcription you already started is automatically linked to your account without losing anything.
How do I upload a file?
From the home page you have two options:
- Local file: drag the file to the upload area or click to select it from your device.
- From URL: paste the link of the platform where the audio or video is hosted (YouTube, for example) and we'll download it for you.
Before clicking "Transcribe" you can set the source language (or leave it on automatic detection) and enable options like speaker recognition or post-transcription translation.
How long does a transcription take?
It depends on the file duration and the options you enable, but in most cases a 30-minute file takes between 1 and 3 minutes. Options like speaker recognition or translation add some time — the page shows you a speed estimate before you start.
Transcription
What file formats do you accept?
The most common audio and video formats: MP3, WAV, M4A, AAC, OGG, OPUS, WMA, FLAC for audio; MP4, MOV, MKV, WebM, AVI, WMV for video. If your file is video, we extract the audio track automatically — you don't need to convert it first.
What's the maximum file size?
It depends on whether you have an active subscription: 2 GB per file and 1 file per batch in guest mode or with a free account; up to 5 GB per file and 3 simultaneous files with any active subscription plan. If your recording is larger, split it into segments or contact us to discuss your case.
What languages do you support?
More than 99 languages, including English, Spanish, French, German, Portuguese, Italian, Mandarin Chinese, Japanese, Arabic, and all major European and Asian languages. By default the system automatically detects the language with over 95% accuracy, but you can select it manually if you know which it is — this improves quality for very short or noisy audio.
Do you recognize multiple speakers?
Yes, with the Recognize speakers option enabled we automatically label who speaks in each segment. It works well for up to about 10 different speakers. You can then rename each one in the editor ("Speaker 1" → "Ana Martinez") and the change applies to the entire transcription.
How accurate is the transcription?
For clean English or Spanish audio with a single speaker we achieve 95–98% accuracy. It decreases with strong accents, multiple overlapping speakers, background noise, or music. Our editor is designed precisely to correct the few remaining errors without having to rewrite entire paragraphs.
Transcription editor
Can I correct the transcription?
Yes. Each completed transcription has an Edit tab where you can modify the text word by word. It's plain text editing, respecting the segment and speaker structure. When you save, a revision is created in the history — you never lose the original version.
What is the revision history?
Every time you save changes in the editor, the previous state is archived as a revision. A side panel shows all revisions with their date, a summary of the change, and a button to restore whichever version you want.
Can I listen to the audio while editing?
Yes. All tabs in the editor (Transcript, Edit, Translate, Summary) share an audio player at the bottom. Click any line with a timestamp and the audio jumps to that exact point. The player doesn't stop when you switch tabs — it keeps playing where it was.
Can I rename the file or the speakers?
Yes to both. The file name can be changed using the pencil icon in the top bar. Speakers are renamed from the Transcript tab — changing "Speaker 1" to "María" updates it throughout the transcription, translation, and summary. Name changes are also recorded in the revision history.
Translation
Can I translate my transcription?
Yes. Once the transcription is complete, go to the Translate tab and choose the target language. We translate segment by segment respecting timestamps and speakers, and you'll see the progress in real time — you don't have to wait for it to finish before you start reading.
How many languages can I translate into?
We support more than 20 common target languages. You can have multiple translations active at once for the same file (English → Spanish and English → French, for example). Each one is managed separately: adding, deleting, or downloading are independent operations.
What does the translation look like?
Two columns side by side: original on the left, translation on the right. The scroll of both columns is synchronized and hovering over a segment highlights the equivalent in the other column. You have a toggle to choose between segments view (line by line with timestamp) or paragraphs view (grouped by speaker change or long pauses).
Can I download the translation?
Yes, in the same formats as the original: TXT, SRT, VTT, and JSON. Each download includes the language code in the filename.
Automatic summary
What does the summary include?
Four blocks generated from the transcription:
- Executive summary — 3–5 bullet points capturing the essentials.
- Chapters — thematic sections with titles and clickable timestamps (click and the audio jumps there).
- Key points — notable quotes or moments with their timestamp.
- By speaker — only if your audio has two or more speakers: speaking time, number of turns, and an individual summary.
How long does it take to generate?
Between 30 seconds and 2 minutes depending on the length of the transcription. It's faster than translation because it processes the full text once instead of segment by segment.
Can I regenerate the summary if I don't like it?
Yes. From the transcription menu you have "Regenerate summary", which deletes the previous one and launches a new one. Useful if you've significantly edited the transcription after the first generation.
Export and download
What formats can I export to?
- TXT — plain text, without timestamps.
- SRT and VTT — standard subtitle formats for video, compatible with YouTube, Premiere, Final Cut, and web players.
- JSON — full structure with segments, timestamps, speakers, and metadata. Ideal if you're going to process it in another program.
- Original audio — direct download of the file you uploaded (useful if you lost it from disk).
Does downloading the audio use my transcriptfy data quota?
No. The audio is served directly from our storage with a temporary signed URL — it doesn't go through our backend. It's fast and doesn't affect your quota.
Can I download files in bulk?
Not yet — each transcription is downloaded individually in the format you choose. It's an improvement we have on our radar.
Account and access
How do I sign up?
Two ways: email + password or Continue with Google. The Google flow only requires one confirmation, it doesn't ask for additional data. In both cases we create your account instantly and take you to your dashboard.
I forgot my password, how do I recover it?
On the login page you'll find "Forgot your password?" — enter your email and we'll send you a link to reset it. The link expires after a short time for security. If it doesn't arrive, check your spam folder.
Can I change the interface language?
Yes, from settings. We support Spanish, English, German, French, and Portuguese. It's independent of the language you're transcribing in — you can have the interface in English and transcribe in Spanish without any issue.
Can I delete a transcription?
Yes, from the menu of each transcription in the dashboard or from the context menu inside the editor. Deleting it removes the text, translations, summary, and the associated original audio. The action is not reversible.
Guest mode
What can I do without signing up?
Transcribe a sample of up to 30 seconds per file. You'll see the resulting text and can decide whether you want to continue with the full file. For that you'll need to sign up — when you do, the sample becomes your first complete transcription without losing any of your progress.
Why do you ask for a verification when uploading?
Because without registration we have no way to prevent automated abuse. We use Cloudflare Turnstile, an invisible or nearly invisible verification that confirms you're a real person without showing you an annoying CAPTCHA in most cases.
How long do you keep my file if I don't sign up?
The audio and transcription in guest mode are deleted 24 hours after they were uploaded. If you want to keep your work beyond that deadline, sign up before it expires — doing so links the transcription to your account and it won't be automatically deleted.
Subscription and payments
What plans are available?
We work with a minutes package model: you choose the package that best fits your monthly volume and pay a per-minute price that decreases with larger packages. The available packages, the per-minute price for each, and the included features are explained in detail on the pricing page. That's where we centralize everything so it's always the up-to-date reference.
How do I pay?
By card through Stripe. We accept Visa, Mastercard, American Express, and European cards with 3D Secure. The charge is recurring (monthly) and you can cancel at any time from settings.
Can I change my package?
Yes, at any time from Settings → Subscription → Manage plan. The details of how changes are applied (immediately or at the end of the cycle) are shown in that same modal before confirming — so you see exactly what you'll pay and when before accepting.
What happens if I cancel?
You keep full access to your package until the end of the already-paid billing cycle. When that cycle ends, the subscription becomes inactive — you don't lose your previous transcriptions, you just stop consuming new minutes until you purchase another package.
Where do I see my payments and account data?
Everything related to your subscription — active package, next charge, payment history, and your account data — is found in Settings → Subscription.
How do I know how many minutes I have left?
On the same Subscription page there's a usage bar showing minutes used versus minutes available in the current cycle. It updates every time you complete a transcription.
Security and privacy
Where are my files stored?
On Cloudflare R2, with encryption at rest and access through temporary signed URLs. The upload from your browser goes directly to storage, without passing through intermediate servers where the file would be exposed.
Do you use my transcriptions to train AI models?
No. Your content is yours — we don't use it for training and we don't share it with third parties beyond the processing needed to generate the transcription, translation, or summary you requested.
How do you protect my account?
Passwords stored with hashing (never in plain text), session cookies with secure and httpOnly flags, and rate limiting on sensitive endpoints. We recommend using a long, unique password — or better yet, sign in with Google and let them handle the second factor.
Can I exercise my GDPR rights?
Yes. You have the right to access, rectification, erasure, portability, and objection to the processing of your data. You'll find the details and contact channels in our GDPR policy and you can write to us to exercise them at any time.
Still have questions?
How can I contact support?
From the contact page you can send us a message. We respond during European business hours — usually the same business day. For issues that are blocking your work, include the transcription ID (you can see it in the URL when you have it open).
Is there an API to integrate transcriptfy into my application?
We don't currently offer a stable public API. If you have a specific integration case write to us from the contact page and we'll study it case by case.
What do I do if I find a bug?
We'd love to know about it. Write to us through the contact page explaining what you were doing, what you expected to happen, and what actually happened. Bugs with steps to reproduce are gold — they go to the top of the queue.