Telegram Voice to Text: Transcribe Voice Messages Free Online

How to convert a Telegram voice message to text

The only slightly fiddly part is getting the voice message out of the chat as a file. Once you have it, the transcriber above does the rest.

1. On Telegram Desktop: right-click the voice message and choose Save As. The file saves as .ogg (Opus audio inside an OGG container).
2. On the phone app: tap the voice message, then use Share or Save to Files to export it. The exported file is usually .ogg or .oga — both work here; if you get .oga, renaming it to .ogg is fine since it is the same container.
3. Open the transcriber at the top of this page and upload the saved file. No conversion to MP3 first — .ogg and .opus go straight in.
4. Leave the language on auto-detect, pick Best accuracy for a voice note, and press transcribe.
5. Copy the punctuated text or download it as a .txt file.

Why transcribe a Telegram voice message instead of just listening

Telegram does have a built-in voice-to-text feature, but on free accounts it is rationed — a small number of transcriptions before it asks you to upgrade to Premium. If you get more voice messages than that allowance covers, an external transcriber is the practical way to keep reading them as text.

Beyond the free-tier cap, text is simply more useful than audio in a lot of situations. You can search a transcript for the one detail you need instead of scrubbing through a three-minute ramble, paste it into notes, quote it in a reply, or read it when you cannot play sound out loud. For long voice notes, reading is far faster than listening at 1x.

This page is built for that task: turning a voice file that came from Telegram into text. It is an independent tool and is not affiliated with or endorsed by Telegram.

Telegram voice notes are OGG/OPUS — and that is handled here

When Telegram records a voice message it encodes it with the Opus codec inside an OGG container. That is a deliberate choice: Opus keeps speech clear at very low bitrates, which is why messaging apps use it. The file you export therefore ends in .ogg (or sometimes .opus or .oga), not .mp3.

Many transcription sites reject that format and force you to convert OGG to MP3 first, which adds a step and re-encodes already-compressed audio. This transcriber decodes .ogg and .opus directly, alongside MP3, WAV, M4A, and AAC, so you upload the Telegram file exactly as it was saved.

Works in 40 languages, detected automatically

Telegram is used heavily across Russian-, Arabic-, Spanish-, and Persian-speaking regions, so voice messages come in every language. The model supports 40 language locales — including English, Russian, Arabic, Spanish, French, German, Portuguese, Ukrainian, Turkish, Hindi, Japanese, Korean, and Mandarin Chinese — with automatic detection, so you do not have to tell it what was spoken.

It also follows a speaker who switches language mid-message, which is common in real chats where a sentence starts in one language and finishes in another. The transcript tracks the switch instead of forcing everything into a single language.

Free limits, stated plainly

The free tier is sized for exactly this use case, and the caps are published here rather than sprung on you after you upload.

What this Telegram voice to text tool does not do

So you do not waste an upload, here is the honest boundary.

• It transcribes a voice file you have already saved. It does not connect to your Telegram account, read your chats, or pull messages automatically — you export the file, then upload it.
• Output is plain punctuated text only — no SRT or VTT subtitles and no word or segment timestamps.
• No speaker diarization. A group voice note with two people comes out as one continuous transcript without speaker labels.
• Audio files only. Video notes are not accepted; extract the audio first.
• One file at a time — no batch uploads and no public API.
• This is an independent site built on NVIDIA's openly released Nemotron model. It is not affiliated with or endorsed by Telegram or NVIDIA.