Potok turns speech into text in 33 languages, including English and Russian.
An auditor who prepares an audit opinion, working papers or a letter to management usually types such texts by hand. Voice input speeds up dictation: Potok (potok.now) inserts text into any field, and the model is trained for Russian speech, including terms like materiality and audit risk. Potok has a WER of 2.1% on our GolosTest benchmark.
An auditor formulates reports, working papers and letters to management precisely, and here it is important not to lose the legal precision of the wording. Is it worth trusting voice input with texts where every word carries weight, and where to start if you decide to try?
What an auditor should check before voice input of documents
Before dictating an auditor's working papers, it is worth going through several points, from the terms of the profession to the confidentiality of information and the recording hotkey itself.
- Terms of the profession. Materiality, audit risk, sampling and similar words are recognized by Potok with a model trained for Russian speech, including terms; for abbreviations and specific names that are needed often, there is a personal dictionary in the Pro plan.
- Audit secrecy. The audit organization, an individual auditor and their employees are obliged to keep confidential the information that makes up audit secrecy (Art. 9 307-FZ, on audit activity). This should be considered before dictating such information into any program: audio and transcripts are stored on servers in Russia.
- Recording hotkey. By default it is Ctrl+Space; if the combination is taken in the work program where the audit opinion is prepared, it is changed in the settings to any other key combination or a side mouse button, and then there is no conflict.
- Text formatting. Punctuation, capital letters and numbers are placed by it automatically, amounts and percentages in the opinion too, without a separate command.
- Internet connection. Without it recognition does not run; if a recording was still made, it stays in the history, and it can be recognized again when the connection returns.
Doctors face a similar question about secrecy and terms, and there is a review for them in the article on dictation for a doctor.
| What | Examples |
|---|---|
| Documents | audit opinion, auditor's working papers, letter to management |
| Terms | materiality, audit risk, sampling |
| 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 |
| Spokenly | macOS, Windows, Linux, iOS | local models (Whisper/Parakeet) are free forever with no word limit; connecting your own API keys (BYOK) is also free | $9.99/mo (Pro, monthly); $99.99/yr (Pro, yearly (≈$8.33/mo)) | not disclosed | hybrid: locally on the device (Whisper/Parakeet, free) or in the cloud (GPT-4o Transcribe, Deepgram Nova 3, Soniox and others, via BYOK or Pro) |
| v2txt.ru | Windows, macOS | Start plan: 2000 words/week | 0 ₽/n/a (all of Pro is free until the end of 2026 (the regular Pro/Premium price after 2026 is not published on the website)) | rubles directly (Visa, Mastercard, Mir cards and T-Bank are accepted) | cloud (Russia) in online mode; offline mode processes on the device |
| Voco (TsRT) and Laitis: legacy voice input programs for Windows | Windows | Voco: unknown, was a paid product with Basic/Professional/Enterprise versions, there was no free version. Laitis: there is a free version ("free, no ads"), plus a separate paid version which, according to the developers, is distinguished by stable operation | — | not disclosed | Voco: judging by the user manual (2022), Voco's speech recognition historically worked locally on the PC (offline, no internet), which set it apart from modern cloud services, but this is not confirmed on the current site, since the product is not found in the current catalog. Laitis: the laitis.ru site states that speech recognition requires an internet connection, that is, cloud processing |
| 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) |
| 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) |
| Voice Access in Windows 11 | Windows | free, built into Windows 11 version 22H2 and newer | — | not disclosed | on the device (offline): speech recognition on the device after a one-time download of the language pack; the internet is needed only for the initial language download |
| Google voice input (Google Docs voice typing + Gboard) | Web (Google Docs, Chrome/Edge/Safari, current versions), Android (Gboard), iOS (Gboard) | free, built into Google Docs and the Gboard keyboard | — | not disclosed | cloud: in Google Docs, speech recognition is handled by the browser, which manages the speech-to-text service and determines speech processing, after which the text is sent to Google Docs (that is, a cloud service of the browser or Google is used, not local processing). In Gboard some advanced voice typing features for certain locales run on the device and are not sent to Google servers by default, but ordinary Gboard voice input is generally cloud-based |
| 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) |
| MacWhisper | macOS | A free version is available ("Free forever", limited functionality), Pro is paid | €64/one-time (Pro, one-time purchase through Gumroad, lifetime updates); $29.99/yr (Pro, App Store version ("Whisper Transcription"), subscription); $99.99/one-time (Pro, App Store version, one-time purchase) | not disclosed | on the device (offline) by default, with local Whisper models; there are optional cloud transcription providers, but not as the main mode |
| 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) |
- Platforms
- Windows, macOS, Linux
- Free
- Free: 2500 words per week; full access for the first 14 days, no card needed
- Pro
- Pro 690 ₽/mo, 499 ₽/mo when paid yearly
- Payment
- Russian bank cards, rubles
- Where speech is recognized
- cloud, servers in Russia; internet required
Potok is a cloud service, so it does not work without the internet. Download: potok.now/download
Where to start in the first week of dictation
The first week with Potok is worth testing on texts that do not fall under audit secrecy: drafts, internal notes, letters without client data. This makes it easier to get used to the hotkey and notice which terms are worth adding to the personal dictionary. The default hotkey is Ctrl+Space, and it can be changed at once in the settings to a convenient combination or a side mouse button if it is already taken in the work program. It is useful in these same days to estimate how many words usually go into one audit opinion or a letter to management, so it is clearer whether the free limit is enough or Pro is needed at once. 2500 words per week are available for free on one device, which is usually enough for testing, while for daily dictation of an auditor's working papers Pro may be needed. A similar review of terms and a dictionary for a profession exists for programmers too, in the article on voice input for programmers (in Russian).
Where it is better to stay on the keyboard
Potok is a cloud service: the voice goes for recognition to servers in Russia, and it does not work without an internet connection. It has no voice commands or voice editing: edits in an audit opinion are made with the keyboard. It also has no offline mode and no phone versions, so you cannot dictate a document away from a computer through it. If a company's internal policy forbids cloud processing of working papers, only the business option fits, with servers inside the company perimeter.
Frequently asked questions
The audit organization, an individual auditor and their employees are obliged to keep confidential the information that makes up audit secrecy (Art. 9 307-FZ). This should be considered before dictating such information into any program: audio and transcripts are stored on servers in Russia.
The model is trained for Russian speech, including terms. Rare abbreviations and specific names are added to the personal dictionary, which is included in the Pro plan.
No: recognition is in the cloud, and without a network connection you cannot dictate text; it will resume only when the connection returns.
Voice input for an auditor. 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 an auditor you dictate into the same field where you normally type, so there is no need to switch between apps.
Voice input for an auditor. 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 an auditor you dictate into the same field where you normally type, so there is no need to switch between apps.
Voice input for an auditor. 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 an auditor you dictate into the same field where you normally type, so there is no need to switch between apps.
Check it in one evening
Install Potok, dictate what you usually type and compare. 14 days of full access, no card needed. Potok is a cloud service, so it does not work without the internet.
Download PotokThe Potok subscription also includes the Telegram bot @potok_now_bot: forward it a voice message or a long meeting recording and it returns the text. Without a subscription the bot gives 60 minutes per month for free.
Primary sources: support.microsoft.com · support.apple.com