Potok transcribes Russian speech; this site has been translated into English for convenience.
Text from a YouTube video is obtained by transcription, not dictation: dictation is entering text by voice in real time, while here the speech that is already in the video has to be recognized. Send the link to the video to the Potok bot (potok.now) @potok_now_bot, and it will transcribe the Russian speech and send back the text; 60 minutes per month are free.
Subtitles for someone else's YouTube video are not dictated. They are obtained by transcription: a program or bot listens to the speech that is already recorded and turns it into text, rather than catching the voice of a person who is speaking right now.
Transcription and dictation are different tasks for YouTube videos
Dictation is when a person speaks and text appears in an input field in real time, whether it is a message, a document or a search bar. It does not work that way with a YouTube video: there is already a finished audio track, and the task is to recognize what is recorded in it, not what is being said now. This is what is called transcription: you give a bot or service a link or a file, not a live voice. The difference between these two tasks in practice is covered in detail in the article "transcription and voice input: what is the difference". For subtitles from a YouTube video you need exactly transcription: the result is text that you can read, copy or format as a subtitle file.
Ways to get text from a YouTube video
There are several ways to get text from a YouTube video, and they are built differently:
- Send the link to the video to the Potok bot @potok_now_bot in Telegram, or paste it into the bot's mini app. Both accept a direct link without downloading the file and recognize Russian speech.
- If the video is in Russian, the retelling feature built into Yandex Browser will put together a short summary of the video right in the browser for free. It is a retelling, not a word-for-word text with subtitles.
- If you already have an audio or video file at hand instead of a link, one of the web services such as Speech2Text.ru, TurboScribe or Happy Scribe will do. They accept an uploaded file, not the address of a page with a video.
A broader comparison of transcription services sets them all side by side: bots that accept a link, and web services for uploaded files.
When the bot will not fit for YouTube subtitles
If the video is not in Russian, the Potok bot will not transcribe it: it has no translation into another language. You also cannot watch a stream and get text at the same time as the broadcast: the bot needs a finished recording, not a connection to a live broadcast. It will not accept a link to Instagram or TikTok, only video and links from YouTube, VK Video and RuTube. And without the internet the bot does not work: it is cloud-based, and recognition runs on Potok's servers in Russia. If after this you need to build not just subtitles but short results of a meeting or a video from the text, the plan of action is the same as for meeting notes from a recording.
Frequently asked questions
Yes, if you just paste the link: the Potok bot and its mini app accept a direct YouTube link without downloading the video to the computer.
How to start
Potok is available for Windows, macOS and Linux. For two weeks everything is open for free, after that 2500 words per week without paying. Potok is a cloud service: speech is recognized on servers in Russia, and it does not work without the internet.
Download PotokYou dictate in the app, and recordings and voice messages are transcribed by @potok_now_bot in Telegram. 60 minutes per month are free; the Potok subscription removes the limit in the bot and in the app.
Primary sources: yandex.ru · vc.ru · speech2text.ru · audio-2-text.ru