Questions people ask about Transcrib Si

Question 1

Is there a free plan, and do I need a card?

Accepted Answer

Free is $0 with 3 files a day, up to 30 minutes per file and 64 MB per file. No card, no trial clock — the limit is per day rather than per month, and the next upload is refused until the next day.

Question 2

How much does the Pro plan cost?

Accepted Answer

Pro is $10 a month when you pay yearly ($120 a year) or $20 a month when you pay monthly. Either way the allowance is the same: fair use of 1,500 minutes of transcribed audio per calendar month, with every export format, API keys and MCP access.

Question 3

How much does the Team plan cost?

Accepted Answer

Team is priced on request, so there is no list price to publish. It carries everything in Pro plus shared workspaces, a shared glossary, SSO, user management, a committed volume written down and priority support.

Question 4

What does the API cost?

Accepted Answer

API usage is metered and billed by volume, separately from any subscription, with no monthly fee. We do not publish a per-hour rate: write to us with your monthly volume and we will send the rate for it.

Question 5

Do paid plans renew automatically?

Accepted Answer

Yes, they renew until you cancel. Yearly billing is charged once for the year and monthly billing every month. A payment that cannot be collected is retried, and then the account moves to the Free limits.

Question 6

Can I cancel at any time?

Accepted Answer

Yes — from the Billing section of the app or by writing to us. You keep the plan until the end of the period you have paid for, and the cancellation is not a request for a refund of the time already used.

Question 7

Do you give refunds?

Accepted Answer

We do not offer refunds for periods already used, except where the law requires it. If we break a promise on our side, we will make it right — write from the address on the account.

Question 8

Can I get an invoice with my company details?

Accepted Answer

Yes. Write from the email address on the account and tell us the name and address to put on the invoice. We will send it from the same inbox you wrote to.

Question 9

Is there an unlimited plan?

Accepted Answer

No plan lifts the published limits — the numbers we publish are the numbers you get. You may not deliberately evade a limit, split one account across several to get more allowance, or re-encode a media library at scale on a consumer plan.

Question 10

How big can one file be?

Accepted Answer

64 MB per file, on every plan including Team. A file over the limit is rejected at upload with a clear error, before any transcription work is done.

Question 11

How long can one recording be?

Accepted Answer

120 minutes (two hours) per file, on every plan. A longer recording has to be split into parts, and the guide for a long recording shows how to cut it and how to put the timecodes back together.

Question 12

How many files can I send on the free plan?

Accepted Answer

Three files a day, each up to 30 minutes. The counter is per day rather than per month, and the next upload is refused until the next day — reaching the limit does not cancel or suspend the account.

Question 13

Why was my file rejected?

Accepted Answer

Either it was over 64 MB or over 120 minutes, or — on the free plan — over 30 minutes. The error message says which one it was, and the widget shows the same message.

Question 14

What is the Pro allowance?

Accepted Answer

Fair use of 1,500 minutes of transcribed audio per calendar month — 25 one-hour meetings, or 12 two-hour sessions. The counter resets on the first day of each month, and going over it returns a clear error without switching the account off.

Question 15

Can I upload a 2 GB recording?

Accepted Answer

Not as one file. The chunked upload API sends the pieces in 8 MB requests, but the assembled file still has to fit 64 MB and 120 minutes, so split the recording first.

Question 16

How do I split a long file?

Accepted Answer

Any audio or video editor that can cut and export will do: cut at a pause, export each part as mp3 or wav, and upload the parts one by one. Keep every part under 120 minutes and 64 MB.

Question 17

Do the timecodes of the split parts line up?

Accepted Answer

Each part is transcribed on its own, so each part starts at 00:00. The API returns the start and the end of every segment, so you add the offset of each part to the timecodes and concatenate the exports.

Question 18

What happens when I cross a limit?

Accepted Answer

The service answers with the same error shape on every endpoint: {"error": {"code": "limit_exceeded", "message": "..."}}, with HTTP 402 or 413. The message says which limit was crossed, and the account stays open.

Question 19

How long can a file be on the free plan?

Accepted Answer

Up to 30 minutes per file on Free, and up to 120 minutes on the paid plans. A free upload of a longer recording is refused with a clear error that names the limit.

Question 20

Which languages do you support?

Accepted Answer

Two: Russian and English, including speech that switches between them inside one sentence. We would rather say that plainly than show a list of a hundred flags.

Question 21

Do you support Russian and English in the same file?

Accepted Answer

Yes. Our engine runs a model profile tuned for Russian speech and for the Russian-English mix, so a meeting that switches language inside a sentence is transcribed in the language it was spoken.

Question 22

How is the language decided?

Accepted Answer

You set it — ru or en — or leave it on automatic detection. The answer says which source was used: user, auto or fallback. The detection took 0.39 s on a short file in our measurement of 21 October 2026.

Question 23

Does the service translate the transcript?

Accepted Answer

No. The transcript is in the language that was spoken — translation is not part of the product today. The API, the MCP server and the plugins are language-neutral, but they transcribe rather than translate.

Question 24

Do you support languages other than Russian and English?

Accepted Answer

No. Two languages today. The pages that compare a language with the market are a comparison of the other services, not a promise from us — our own engine covers Russian and English and we do not count German in our language list.

Question 25

Is there a Russian-language interface?

Accepted Answer

The interface is in English today. The service handles Russian audio, and the API, the MCP server and the plugins are language-neutral — they work the same way whatever the language of the recording.

Question 26

Can I dictate in Russian and get the text?

Accepted Answer

Yes — a dictation is just a small audio file, so it goes through the same pipeline and comes back with the same timecodes. There is a separate guide on transcribing voice memos.

Question 27

Does it work for a Russian-language meeting?

Accepted Answer

Yes. A Russian or mixed Russian-English file is processed on our own engine, which is tuned for Russian speech. Language detection took 0.39 s on a short file in our own measurement of 21 October 2026.

Question 28

Which export formats do you get?

Accepted Answer

Six: TXT, Markdown, SRT, VTT, CSV and DOCX. DOCX is a Pro feature — on the free plan the other five are there, with timestamps and speaker labels.

Question 29

Which export format should I choose for subtitles?

Accepted Answer

SRT for video subtitles, VTT for web players. Both carry the start and the end time of every segment, so the subtitles line up with the audio. The file is named after the recording.

Question 30

When should I use the CSV export?

Accepted Answer

CSV is one row per segment, with the time and the speaker, so it goes straight into a spreadsheet or into a script that reads the timings. The plain text export is the same transcript without the columns.

Question 31

Can I get the transcript as a Word document?

Accepted Answer

Yes. DOCX is one of the six exports and a Pro feature: a Word document with the transcript, for reviews and hand-offs. On the free plan the other five formats are available.

Question 32

What is the difference between TXT and Markdown?

Accepted Answer

TXT is plain text, one line per segment, for a document or a ticket. Markdown is the same text as headings and lists, for notes, wikis and chat.

Question 33

Which file formats can I upload?

Accepted Answer

Audio and video containers: mp3, wav, m4a, mp4, webm, ogg, flac, aac, wma and mov. One file at a time, up to 64 MB and 120 minutes.

Question 34

Can I get the transcript in a chosen format from the API?

Accepted Answer

Yes: GET /v1/transcripts/{id}/export?format=txt|md|srt|vtt|csv|docx returns the file with the right media type. An unknown format is rejected with invalid_request and the list of the supported ones.

Question 35

Can I fix a word before exporting?

Accepted Answer

Yes. Correct the text in the editor, press Save — the original audio is not changed — and then export. The API reads the same transcript back, line by line, with the timecodes intact.

Question 36

Do you remove filler words from the transcript?

Accepted Answer

No. The transcript is what was said. Editing it takes one click in the editor, and the exports follow the edited text.

Question 37

What happens to my audio?

Accepted Answer

It is used to produce the transcript and to run the service. We do not sell it and we do not train models on it for other customers. Delete a transcript in the app and the audio goes with it.

Question 38

Is my file deleted after transcription?

Accepted Answer

Not automatically: files and transcripts stay in your account until you delete them there, and on the Free plan transcripts are kept for 7 days. A deleted file leaves the live system immediately and the encrypted backups within 30 days.

Question 39

Where is my file processed?

Accepted Answer

A Russian or a mixed Russian-English file is processed on our own transcription infrastructure, which is outside the United States. A file that is entirely in English is passed to a third-party cloud speech provider in the United States when that fallback is enabled on the server.

Question 40

Who else receives my data?

Accepted Answer

Only the services that do the job: our own transcription infrastructure, the cloud fallback provider for clean English, Stripe for payments and invoices, and our hosting and email providers under written data-processing terms. Nobody else, and we do not sell your data.

Question 41

Do you store my card details?

Accepted Answer

No. Stripe handles card numbers, CVVs and bank details; we never receive or store them. We keep only the Stripe customer identifier, the plan, the subscription status and the invoices.

Question 42

Do you train models on my audio or my transcripts?

Accepted Answer

No. We do not use your audio or your transcripts to train models for other customers, and we do not sell your data to anyone. The content data is used to produce the transcript you asked for and to show it back to you.

Question 43

Do you track me with cookies or pixels?

Accepted Answer

The site sets no advertising cookies and no tracking pixels, and we run no advertising and no cross-site tracking. It uses your browser's local storage for the session token and for the light or dark theme choice.

Question 44

Can I get a copy of my data or close my account?

Accepted Answer

Yes: edit or delete a transcript in the app, or write to support@transcrib.si with the subject "Close account" — the account data is deleted within 30 days. A copy of your data on request is answered within 30 days.

Question 45

Do you sign data-processing agreements?

Accepted Answer

Yes, for Team and API customers. Ask and we will send the current version — the agreement is written for the volume and the setup, not as a standard paper.

Question 46

Can you send me my password?

Accepted Answer

No, and nobody else can read it either: passwords are stored only as salted hashes. API keys are stored the same way — the full value is shown once, right after you create it, and never again.

Question 47

What is the API base URL?

Accepted Answer

https://api.transcrib.si, versioned under /v1. The same host serves the MCP server at /mcp, so one deployment carries the REST API, the OAuth endpoints and the MCP tools.

Question 48

How do I authenticate against the API?

Accepted Answer

Send an API key in the X-API-Key header, or a session token as Authorization: Bearer. Both work on the same endpoints, and a key can be revoked from the app at any moment — revocation takes effect immediately.

Question 49

How do I send a file to the API?

Accepted Answer

One multipart POST to /v1/transcripts with the file, a language field (ru, en or auto) and a diarize field (true or false). The answer is {"id": ..., "status": "queued"}; read the transcript back with GET /v1/transcripts/{id} until the status is done.

Question 50

Is there an SDK?

Accepted Answer

No SDK — it is a plain REST API with one file-upload endpoint, multipart. requests or httpx is enough, as the example on the API page shows.

Question 51

What does an API error look like?

Accepted Answer

Every endpoint answers with the same shape: {"error": {"code": "...", "message": "..."}}, so one handler covers the whole API. The codes are unauthorized, forbidden, not_found, limit_exceeded, invalid_request, provider_error, payment_required, billing_disabled and not_ready.

Question 52

Where do I get an API key?

Accepted Answer

In the app, on the Pro or Team plan. The value is shown exactly once at creation and we store only its hash, so it cannot be shown again — copy it at once, and revoke it if it leaks.

Question 53

Is there an MCP server?

Accepted Answer

Yes, the same host serves it at /mcp with four tools: transcribe_audio, summarize_transcript, extract_actions and list_transcripts. The Claude and ChatGPT plugins use that same server.

Question 54

Do I need an API key for MCP?

Accepted Answer

No: the authorisation is OAuth 2.1 with PKCE, so the user's own session is what allows the call. An API key also works for a server-to-server call.

Question 55

Does the assistant read my chat history?

Accepted Answer

No. The plugin does not read the chat history and does not pass chat context to the service: only the audio file, its name and the language you specify are sent over TLS. The server does not call other services either.

Question 56

Can I upload a file in parts through the API?

Accepted Answer

Yes: POST /v1/uploads/init returns an upload id and the chunk size of 8 MB, then PUT /v1/uploads/{id}/chunk?index=N for each piece and POST /v1/uploads/{id}/complete to assemble it. The limits on the finished file stay 64 MB and 120 minutes.

Question 57

How do I check how much I have used?

Accepted Answer

GET /v1/usage returns the plan, the period, the minutes and the files used; GET /v1/me returns the current user, the plan and the usage. The same numbers are in the Usage section of the app.

Question 58

How can I check that the transcription engine is alive?

Accepted Answer

GET /v1/asr/endpoints answers with the configured and the discovered engine address, whether the engine was discovered and healthy, and the capabilities: transcribe, timed, language, diarize and speakers. The app shows the same status under Service status.

Question 59

Is there a webhook when a transcript is ready?

Accepted Answer

No. The endpoint table of the API has no transcript webhook: you read the transcript with GET /v1/transcripts/{id} until the status is done.

Question 60

What does the Stripe webhook in the API do?

Accepted Answer

POST /v1/stripe/webhook receives the events of Stripe and applies them to the account: it is the billing side of the service. It is not a transcript callback — no webhook fires when a transcript is ready.

Question 61

Is there a live stream or a websocket?

Accepted Answer

No. The service takes files: you send the file and read the transcript when it is done. There is no live transcription in the product.

Question 62

What does the API return for a transcript?

Accepted Answer

The text, the status, the segments with start_s and end_s, the speaker labels, the language and the duration. Every segment carries its start and its end, and the exports carry them too.

Question 63

Can I delete a transcript through the API?

Accepted Answer

Yes: DELETE /v1/transcripts/{id} removes the transcript and the file with it. A deleted file leaves the live system immediately and the encrypted backups within 30 days.

Question 64

Do you have a Telegram bot?

Accepted Answer

The bot is announced on this site, but its handle and its link are placeholders until it answers — the site carries it with the status "placeholder". Write to us if you want to be told when it is live.