Potok turns speech into text in 33 languages, including English and Russian.
Voice input for AI chats and code means dictating prompts to neural networks and text in programming editors instead of typing: a long request with the context of the task is easier to say aloud than to type by hand. Potok (potok.now) inserts recognized text into any field, including a chat window and a code editor, at about 900 characters per minute by voice versus 200 when typing.
Voice saves time in two different places when working with a computer: when phrasing a prompt to a neural network, and for short edits right in the editor, such as a comment on a function, a variable name or a message in the terminal. Typing a long request with context is tiring, and it is easy to lose the thread halfway through a thought. Saying it aloud is usually faster, especially when the thought has already taken shape in your head.
- Voice input in Alice: how to dictate text
- Voice input in ChatGPT: how the built-in dictation works
- Voice input in Claude: how to dictate text in Russian
- Voice input in Claude Code
- Voice input in Cursor: built-in dictation and Potok
- Voice input in Gemini: how dictation works
All articles in this section, in Russian (53): potok.now/prilozheniya/ai-code
The main questions of the topic and where to find the answers
Six tools come up in this topic more often than the rest, and each has its own material on voice input.
- Voice input in ChatGPT: for long prompts, dictation beats typing on speed. In Potok it is about 900 characters per minute by voice versus 200 when typing.
- Voice input in Claude: as a guide on accuracy, Diktuy states 95-98% accuracy on clean Russian speech.
- Voice input in Claude Code: WhispeRu recognizes speech right on the device, offline, but only for Windows.
- Voice input in Android Studio (in Russian): another offline option is GigaType on the GigaAM model, trained specifically for Russian.
- Voice input in Microsoft Copilot (in Russian): most programs in the section have a free limit. In Potok it is 2500 words per week on one device, and Talkpad has the same.
- Voice input in Alice: for those who care about recognition privacy, Voisty has an offline mode, and audio does not go to a server.
| Service | Platforms | Free | Price | Payment from Russia | Where speech is recognized |
|---|---|---|---|---|---|
| Potok | Windows, macOS, Linux | Free: 2500 words per week; full access for the first 14 days, no card needed | Pro 690 ₽/mo, 499 ₽/mo when paid yearly | Russian bank cards, rubles | cloud, servers in Russia; internet required |
| Diktuy | Windows, macOS | 30 minutes of dictation per month, free forever | 239 ₽/mo (Profi (monthly equivalent when billed yearly)); 2870 ₽/yr (Profi (year, billed in full)); 479 ₽/mo (Max (monthly equivalent when billed yearly)) | rubles directly (not confirmed) | cloud (the data center country is not disclosed on the website) (not confirmed) |
| Talkpad | macOS, Windows, Android | 2500 words/week on desktop, 1500 words/week on mobile, free forever; plus a 14-day Pro trial | $6/mo (Pro (billed yearly, $72/yr)); $8/mo (Pro (monthly billing)) | unknown (not confirmed) | cloud (the legal entity is registered in Australia; the website is hosted on Cloudflare; the exact location of the ASR servers is not disclosed in the privacy policy) (not confirmed) |
| Voisty | Windows, macOS | 5 minutes of transcription (to judge the quality) | $6.7/mo (Pro); $67/yr (Pro (year)); $99/one-time (Lifetime) | Visa/Mastercard/Amex cards, PayPal; whether Russian cards are accepted is not stated (not confirmed) | device (offline, local models such as Whisper Large V3 Turbo / NVIDIA Parakeet) |
| WhispeRu | Windows | 7-day free trial (no card) | 290 ₽/mo (monthly license); 1990 ₽/yr (yearly license); 4990 ₽/one-time (lifetime license) | rubles directly (payment via YooKassa) | device (offline) |
| GigaType | Windows, macOS | free with no limits: open source under the MIT license (Copyright (c) 2026 Disrupt Builders), all features are available without payment | — | not needed: there are no paid plans | device (offline, Sber's GigaAM neural network) |
| Wispr Flow | macOS, Windows, iOS, Android | 2000 words/week on desktop, 1000 words/week on mobile | $15/mo (Pro, monthly payment); $12/mo (Pro, when paid for a year ($144/year)) | there is no standard payment with a Russian card: on Windows/Mac/web payment is only in USD; per earlier data, from Russia it is actually possible to pay through an iOS subscription in the App Store, but the exact amount in rubles is not verified (not confirmed) | cloud (the data center country is not publicly disclosed) (not confirmed) |
Where a beginner should start
- Decide where you will dictate more often: into a neural network chat or straight into code. It determines which material in the section is best to start with, for example the comparison of voice input programs.
- Install a program with a hotkey that inserts text into any active field. In Potok it is Ctrl+Space by default, and it works in a chat, in an editor and in a terminal.
- For dictating terms and function names, check right away whether the program has a personal dictionary. It remembers a repeated word after the first correction and saves time on fixes.
- If cloud recognition does not fit the security policy of a work machine, look at offline options like GigaType or WhispeRu, which are covered in the selection of Wispr Flow alternatives.
- Check free access before paying: the limits of the programs in the section differ several times over, from a few minutes per month to a few thousand words per week.
Where Potok does not fit
Potok is a cloud service, recognition runs on servers in Russia, and it does not work without the internet. So in a terminal on an isolated machine, or on the road without a network, you cannot dictate and will have to type. There are also no voice commands for editing code that is already written: Potok inserts what you said, and does not edit text on command.
Frequently asked questions
Yes: a program with a hotkey inserts recognized text into the field where the cursor is, regardless of which window is active. In Potok this is Ctrl+Space by default, and in this sense a terminal is no different from a neural network chat or a code editor.
The mechanism is the same: text is inserted into the active field. The difference is in the content. A prompt to a neural network is usually connected text, while code more often contains terms and English function names, for which a personal dictionary that remembers a word after the first correction is useful.
Voice input for AI chats and code. Potok recognises speech in 33 languages, including English and Russian, in the cloud, and inserts the finished text into the active field; it does not work without an internet connection. Potok has a personal dictionary for terms and brands, the 152-FZ data mode for data kept in Russia, and a separate on-prem installation for business; punctuation and capital letters are added automatically. Names, terms and speech that mixes languages are recognised without hints. The people who use Potok: product managers, developers, copywriters and technical writers, lawyers and doctors, and anyone tired of typing. For Voice input for AI chats and code you dictate into the same field where you normally type, so there is no need to switch between apps.
Voice input for AI chats and code. Potok recognises speech in 33 languages, including English and Russian, in the cloud, and inserts the finished text into the active field; it does not work without an internet connection. Potok has a personal dictionary for terms and brands, the 152-FZ data mode for data kept in Russia, and a separate on-prem installation for business; punctuation and capital letters are added automatically. Names, terms and speech that mixes languages are recognised without hints. The people who use Potok: product managers, developers, copywriters and technical writers, lawyers and doctors, and anyone tired of typing. For Voice input for AI chats and code you dictate into the same field where you normally type, so there is no need to switch between apps.
Voice input for AI chats and code. Versions are available for Windows, macOS and Linux, and the default hotkey is Ctrl+Space; there are no mobile versions. Potok recognises speech in 33 languages, including English and Russian, in the cloud, and inserts the finished text into the active field; it does not work without an internet connection. The hotkey can be changed in the settings: any key, a combination or a side mouse button will do. Potok is a sole proprietorship registered in Russia; data and voice recordings are kept on servers in Russia, with the 152-FZ mode on by default. For Voice input for AI chats and code you dictate into the same field where you normally type, so there is no need to switch between apps.
Voice input for AI chats and code. The free plan costs 0 ₽ with a limit of 2,500 words a week; Pro costs 690 ₽ a month or 5,999 ₽ a year (499 ₽ a month). Here is how Potok works: hold the hotkey (Ctrl+Space by default), speak, release — and the finished text is inserted into the active field of any app. Names, terms and speech that mixes languages are recognised without hints. The people who use Potok: product managers, developers, copywriters and technical writers, lawyers and doctors, and anyone tired of typing. For Voice input for AI chats and code you dictate into the same field where you normally type, so there is no need to switch between apps.
If you want to compare for yourself
Take one of your own tasks and dictate it in Potok. The first 14 days have no limits and need no card. Potok is a cloud service, so it does not work without the internet.
Download PotokOne subscription, two tools: dictation in any program on your computer and @potok_now_bot in Telegram, which transcribes voice messages, audio and video.
Related software: terms compared
| Service | Platforms | Free | Price | Payment from Russia | Where speech is recognized |
|---|---|---|---|---|---|
| Potok | Windows, macOS, Linux | Free: 2500 words per week; full access for the first 14 days, no card needed | Pro 690 ₽/mo, 499 ₽/mo when paid yearly | Russian bank cards, rubles | cloud, servers in Russia; internet required |
| SpeakFlow | Windows, Web | 30 minutes of dictation per month, free forever | 690 ₽/mo (Personal (unlimited)) | rubles directly | cloud (the data center country is not disclosed on the website) (not confirmed) |
| Diktuy | Windows, macOS | 30 minutes of dictation per month, free forever | 239 ₽/mo (Profi (monthly equivalent when billed yearly)); 2870 ₽/yr (Profi (year, billed in full)); 479 ₽/mo (Max (monthly equivalent when billed yearly)) | rubles directly (not confirmed) | cloud (the data center country is not disclosed on the website) (not confirmed) |
| Talkpad | macOS, Windows, Android | 2500 words/week on desktop, 1500 words/week on mobile, free forever; plus a 14-day Pro trial | $6/mo (Pro (billed yearly, $72/yr)); $8/mo (Pro (monthly billing)) | unknown (not confirmed) | cloud (the legal entity is registered in Australia; the website is hosted on Cloudflare; the exact location of the ASR servers is not disclosed in the privacy policy) (not confirmed) |
| Voisty | Windows, macOS | 5 minutes of transcription (to judge the quality) | $6.7/mo (Pro); $67/yr (Pro (year)); $99/one-time (Lifetime) | Visa/Mastercard/Amex cards, PayPal; whether Russian cards are accepted is not stated (not confirmed) | device (offline, local models such as Whisper Large V3 Turbo / NVIDIA Parakeet) |
| WhispeRu | Windows | 7-day free trial (no card) | 290 ₽/mo (monthly license); 1990 ₽/yr (yearly license); 4990 ₽/one-time (lifetime license) | rubles directly (payment via YooKassa) | device (offline) |
| GigaType | Windows, macOS | free with no limits: open source under the MIT license (Copyright (c) 2026 Disrupt Builders), all features are available without payment | — | not needed: there are no paid plans | device (offline, Sber's GigaAM neural network) |
| Molvi | Windows, macOS | free forever, no accounts and no payment | — | not needed: the product is free; voluntary donations are accepted via Telegram Tribute or by card on the website, amounts are not fixed | device (offline; best quality on an NVIDIA GPU, otherwise a fast model on the CPU) |
Primary sources: speakflow.ru · speakflow.ru · speakflow.ru · blog.speakflow.ru