How to turn a video into text with a bot in Telegram

Potok transcribes Russian speech; this site has been translated into English for convenience.

A video is turned into text in Telegram by the Potok bot @potok_now_bot: send a file or a link to YouTube, VK Video and RuTube, and a short transcript arrives as text, a long one as a file, 60 minutes per month for free. Video notes and voice messages in Telegram have their own transcription button on the message, limited without Premium. For typing your own text by voice there is the Potok app.

A video that needs to be turned into text most often comes as a file, a YouTube link or a video note in Telegram. Video notes and voice messages in Telegram itself have a transcription button on the message, and for other videos you need a separate bot, for example the Potok Telegram bot @potok_now_bot. If a voice message came instead of a video, the steps are the same.

How to get text from a video with a bot in Telegram

1. Open the bot @potok_now_bot in private messages and press Start (/start): the bot accepts video files and links only in a private chat, not in groups.
2. Send a video file (it can be as a document) or paste a link to YouTube, VK Video, RuTube or a direct link to a file.
3. Wait for the reply: a short transcript arrives as text right in the chat, a long one as a separate file; the file format (TXT or DOCX) is chosen in advance in the settings.
4. Under the text the buttons Summary, Tasks and deadlines, Ask and Other format appear: you can get a short digest, a list of tasks by speaker, an answer to your own question about the recording, or the same text in another format.
5. Timecodes [mm:ss] and speaker markup (Speaker 1, Speaker 2) can be turned on in advance in the bot settings (/settings); by default they are off.

What you get after transcribing a video

• A short recording as text right in the chat, without a separate file.
• A long recording as a file in the chosen format (TXT or DOCX).
• Proofread text: punctuation and capital letters are placed, obvious recognition errors are fixed, but the text is not shortened or retold.
• With the settings turned on, timecodes and speaker labels; the bot does not know names, only the numbers Speaker 1, Speaker 2.
• Buttons under the text: a short summary, a list of tasks by speaker, an answer to your own question through Ask, export in one of six formats: TXT, DOCX, Markdown, PDF, SRT or VTT (the last two are subtitle files).
• A finished transcript can be opened again in the My recordings section. A general overview of the categories of services is in the article audio to text transcription: a comparison of services.

When the bot cannot handle a video

The bot will not help with a video that is not in Russian: it recognizes only Russian speech, and there is no translation into another language. It also does not tell speakers apart by name: only Speaker 1, Speaker 2 in the order of the lines. The bot will not pick up a video recording during a call in Zoom or Telemost by itself: the recording must first be saved in the program itself and forwarded as a file. The bot will not accept a link to Instagram or TikTok. And most important: the Potok bot is a cloud one, recognition runs on servers in Russia, and it does not work without the internet.

Frequently asked questions

Primary sources: core.telegram.org · www.ixbt.com · developers.sber.ru · t.me