Free online tool
OGG to Text: Free Online Converter for Voice Notes and Recordings
Upload an OGG or OPUS file below and get clean, punctuated text in seconds. The transcriber runs NVIDIA's Nemotron 3.5 ASR model, detects 40 languages automatically, and needs no account — 3 free transcriptions a day, up to 5 minutes and 20 MB per file. That covers almost every voice note from WhatsApp, Telegram, or Discord.
Your transcript will appear here with punctuation and capitalization.
How to convert OGG to text online
The transcriber at the top of this page handles the whole job. No account, no email, no software install.
- Click the upload area above and pick your file. The tool accepts .ogg and .opus directly, alongside MP3, WAV, M4A, and AAC.
- Leave the language on auto-detect, or choose one of the 40 supported locales if you already know what was spoken.
- Pick a speed mode. Best accuracy is the sensible default for voice notes; the faster modes trade a little precision for quicker turnaround.
- Press transcribe and wait a few moments while the Nemotron 3.5 ASR model processes the audio.
- Copy the punctuated, capitalized text, or download it as a .txt file.
What an OGG file actually is
OGG is a container, not a codec. The .ogg wrapper can hold different kinds of compressed audio — most often Vorbis in older files and Opus in newer ones. That distinction clears up a common confusion: a file ending in .opus is Opus audio sitting inside the same OGG container. Same box, different label.
You meet OGG files everywhere voice gets recorded cheaply and efficiently: WhatsApp and Telegram voice messages, Discord recordings, Audacity exports, Linux screen recorders, game clips. Opus in particular is exceptional at speech — it keeps voices intelligible at very low bitrates, which is exactly why messaging apps chose it.
The catch is that plenty of transcription tools still reject the format and make you convert OGG to MP3 first. This converter decodes OGG and OPUS natively, so you skip the extra step and the quality loss that comes with re-encoding compressed audio.
Transcribe WhatsApp, Telegram, and Discord voice notes
Voice notes are the hidden reason most people need an OGG to text converter. Getting the file out of the chat app is the only fiddly part.
- WhatsApp: open the chat, long-press the voice message, and share or forward it to your email, your Files app, or anywhere that saves it. The exported file usually ends in .opus, sometimes .ogg — both work here.
- Telegram: on desktop, right-click the voice message and choose Save As. Telegram voice messages are Opus audio in an OGG container too.
- Discord and other apps: download the recording from the message; it is typically a plain .ogg file.
- Upload the file exactly as it was saved. No renaming, no converting. If some app hands you a rarer extension such as .oga, renaming it to .ogg works — it is the same container.
- Hit transcribe. Voice notes rarely run past five minutes, so the free tier fits this job comfortably.
40 languages, detected automatically
The model supports 40 language locales, including English, Spanish, French, German, Portuguese, Japanese, Korean, Mandarin Chinese, Hindi, Arabic, Russian, Vietnamese, Thai, Turkish, Ukrainian, and Polish. Leave the selector on auto-detect and it identifies the spoken language on its own.
It also follows a speaker who switches language mid-recording — common in real voice notes, where a sentence starts in English and finishes in Spanish. The transcript tracks the switch instead of forcing everything into one language.
Free limits, stated plainly
The free tier is built for exactly the voice-note use case, and the caps are published up front rather than discovered at checkout.
| Limit | Free | Pro |
|---|---|---|
| Price | Free, no account | $9 per month |
| Allowance | 3 transcriptions per day per connection | 300 audio minutes per month |
| Per-file cap | 5 minutes and 20 MB | 5 minutes and 20 MB |
| Access | Works instantly in the browser | Creem license key, up to 3 browsers |
Four speed modes, one honest trade-off
Nemotron 3.5 ASR is a 0.6B-parameter cache-aware FastConformer-RNNT streaming model. The mode selector changes the audio chunk size the model processes at once: bigger chunks give it more context and better accuracy, smaller chunks return text sooner.
| Mode | Chunk size | Use it for |
|---|---|---|
| Best accuracy | 1.12 s | Voice notes, interviews, anything you will read closely |
| Balanced | 0.56 s | Everyday transcription, a good default |
| Faster | 0.32 s | Quick drafts where speed matters more |
| Lowest latency | 0.08 s | Fastest turnaround, lightest context |
What this OGG to text tool does not do
Some tools ranking for this search promise everything. Here is the honest boundary, so you do not waste an upload.
- Output is plain punctuated text only — no SRT or VTT subtitle export, no word or segment timestamps. If you need subtitles, read the audio to SRT guide linked below before uploading.
- No speaker diarization. Two voices in one recording come out as one continuous transcript.
- Audio files only. Video files are not accepted; extract the audio track first.
- One file at a time — no batch uploads and no public API.
- Transcription runs after the upload or recording finishes. It is not a live caption overlay.
- This is an independent site built on NVIDIA's openly released Nemotron model; it is not affiliated with NVIDIA.
Frequently asked questions
Can I convert OGG to text for free?
Yes. You get 3 free transcriptions per day per connection, no account required. Each file can be up to 5 minutes and 20 MB, which covers nearly every voice note. Pro removes the daily cap with 300 audio minutes per month for $9.
Do you support .opus files from WhatsApp?
Yes. WhatsApp voice notes usually export with a .opus extension — that is Opus audio inside the same OGG container. Upload the .opus file directly; no renaming or conversion needed.
Can I get subtitles or an SRT file from an OGG recording?
No. This tool outputs plain punctuated text without timestamps, and a valid SRT file requires accurate segment timing. If subtitles are the real goal, see the audio to SRT guide on this site before uploading.
What languages does the OGG to text converter support?
40 language locales with automatic detection, including English, Spanish, French, German, Japanese, Korean, Mandarin Chinese, Hindi, Arabic, and Russian. It also follows mid-recording language switches.
Does it separate different speakers?
No. There is no speaker diarization, so a conversation is transcribed as one continuous text without speaker labels.
How accurate is the transcription?
It runs NVIDIA's Nemotron 3.5 ASR streaming model. Accuracy depends mostly on your audio: a clear voice note recorded close to the microphone transcribes very well, while heavy background noise degrades any ASR model. Best accuracy mode gives the model the most context per chunk.
My recording is longer than 5 minutes. What are my options?
Every upload is capped at 5 minutes and 20 MB, on both free and Pro. Split a longer recording into parts with any audio editor and transcribe them in sequence — Pro's 300 monthly minutes make multi-part jobs practical.