To convert an MP3 to text, send the file to the Potok bot on Telegram. The transcript comes back in the chat, or as a file if it is long. The bot accepts files up to 2 GB, which is about 34 hours of MP3 at 128 kbps. Export TXT, MD, DOCX, PDF, SRT or VTT.
Transcribe a recording
Send your voice message, audio or video to @potok_now_bot on Telegram and the text comes back in the chat.
Open @potok_now_bot in Telegram
Files up to 2 GB · export to TXT, MD, DOCX, PDF, SRT and VTT · speaker labels and timestamps · about 30 languages
Key facts
- File size matters more than length: an MP3 at 128 kbps weighs 0.96 MB a minute and 57.6 MB an hour.
- Send the original recording. A copy that was compressed again has already lost detail the recogniser needs.
- Speaker labels and timestamps are optional and appear as Speaker 1, Speaker 2, never as real names.
- A file above the usual chat limit goes through the Upload button of the bot's mini app.
- The text stays in the language of the recording and is not translated.
What an MP3 weighs
MP3 is a lossy format: it makes files small by discarding detail that most listeners do not notice. The size depends on the bitrate, the amount of data used for each second of sound. A voice recording is often saved at 64 to 128 kilobits per second. The table gives megabytes per minute and hour, and how many hours fit into the 2 GB limit.
| Format and quality | kbps | MB per minute | MB per hour | Hours that fit in 2 GB |
|---|---|---|---|---|
| MP3 or AAC, low voice quality | 64 | 0.48 | 28.8 | 69.4 |
| MP3 or AAC, voice | 96 | 0.72 | 43.2 | 46.3 |
| MP3, common default | 128 | 0.96 | 57.6 | 34.7 |
| MP3 or AAC, music | 192 | 1.44 | 86.4 | 23.1 |
| MP3, high | 320 | 2.40 | 144.0 | 13.9 |
| Opus, voice | 24 | 0.18 | 10.8 | 185.2 |
| WAV, 16 kHz mono 16-bit | 256 | 1.92 | 115.2 | 17.4 |
| WAV, 44.1 kHz stereo 16-bit (CD quality) | 1411.2 | 10.58 | 635.0 | 3.1 |
Sizes use decimal units (1 MB = 1,000,000 bytes). MB per minute = kbps × 60 ÷ 8 ÷ 1000. This is a calculation, not a measurement of any file.
How to convert an MP3 to text
Five steps from file to document:
- Open the bot. Open @potok_now_bot in Telegram and press Start.
- Send the MP3. Attach it as an audio file or as a document. For a large file, up to 2 GB, use the Upload button of the bot's mini app.
- Switch on speaker labels. Do this when more than one person talks. Timestamps can be switched on in the same place.
- Wait for the text. A short transcript arrives in the chat. A long one arrives as a file.
- Export. Save the transcript as TXT, MD, DOCX, PDF, SRT or VTT.
What the result looks like
With speaker labels and timestamps on, each passage starts with a time and a speaker number. The example below is text we wrote to show the layout.
[00:00] Speaker 1: Hi Sam, it's Ana from the venue.
[00:03] Speaker 1: The big room is open on Thursday from ten to one, but the projector is booked.
[00:09] Speaker 1: Call me back and tell me if the small room works for you.
Sample text written by us to show the layout of a transcript. It is not the output of the recognition engine and not a real recording.
| Feature | Details |
|---|---|
| Where it works | Telegram bot @potok_now_bot; large files go through the bot's mini app |
| File size | Up to 2 GB |
| Export | TXT, MD, DOCX, PDF, SRT, VTT |
| Speakers and time | Speaker labels and timestamps can be switched on |
| Languages | About 30 languages with good published results; a few more with warnings |
| Translation | The transcript is not translated into another language |
| Dictation | A separate Potok desktop app lets you dictate text into any program; it recognises Russian speech |
Common mistakes
| Mistake | What happens | Fix |
|---|---|---|
| Sending an MP3 that was compressed twice | Quiet words and names are already damaged before recognition starts, so they come out wrong. | Send the original recording. If you must export again, use 96 to 128 kbps. |
| Judging a file by minutes instead of megabytes | A two-hour MP3 at 320 kbps weighs about 288 MB, while two hours of CD-quality WAV is over 1 GB. | Read the size from the table above or use the audio file size calculator. |
| Leaving speaker labels off for a call with several people | The text runs together and you have to work out who said what. | Switch labels on before you send the file. |
| Expecting the labels to carry names | The transcript says Speaker 1 and Speaker 2, because the bot does not know who the people are. | Replace the labels with names in the exported DOCX or TXT. |
| Quoting the transcript without listening | Names, numbers and overlapping voices are where errors cluster. | Play the audio next to the text and correct those places first. |
When you need subtitles
An SRT or VTT export gives you the same text split into short lines with times. Read about the SRT file format if you want to edit the lines by hand. If your MP3 is a voice message from a messenger, send it straight from the chat: see Telegram voice messages. For a recording with several voices, the interview transcription guide covers labels and quotes.
Frequently asked questions
Can I transcribe an MP3 file for free?
This page does not quote prices or limits. Open the bot to see what is available to you.
Does the length of the MP3 matter?
The size of the file matters more than the length. Files up to 2 GB are supported, and the table on this page shows how many hours of MP3 fit at common bitrates.
Is a lower MP3 bitrate bad for transcription?
A very low bitrate removes detail and can make speech harder to recognise. For speech, 64 to 128 kbps is a common choice. If you still have the original recording, send that.
Can I get subtitles from an MP3?
Yes. Export the transcript as SRT or VTT to get subtitle files with a time for every line.
Does Potok translate the text from my MP3?
No. The text stays in the language of the recording.
Sources
Each source was opened by hand on the date shown. Pages change, so open the source if a detail matters to you.
- MP3 (Wikipedia), Wikipedia · checked
- Bit rate (Wikipedia), Wikipedia · checked
- Megabyte (Wikipedia), Wikipedia · checked
Try it on your own recording
Send your voice message, audio or video to @potok_now_bot on Telegram and the text comes back in the chat.
Open @potok_now_bot in Telegram