Does it work with WhatsApp voice notes?

Yes. Export the message (long-press → Share) and drop the .opus file here — no conversion needed first.

Brazilian or European Portuguese?

Both. The model handles the two variants and their accents; you do not need to choose a country.

Do I have to pick the language?

No. On the main transcription page the spoken language is detected automatically. This page simply starts with Portuguese already selected, which saves a step and removes any doubt when the audio is noisy.

Which model should I pick for Portuguese?

Portuguese is well covered by Whisper, so Fast is fine for a clear voice note and Accurate for interviews and meetings. Switch to Best for noisy recordings, strong regional accents or when every word matters.

Are accents and punctuation included in the transcript?

Yes. The model writes standard Portuguese with accents, cedillas, tildes and punctuation, so the text can be pasted into a document without clean-up. Review proper names and numbers, which are the usual weak points.

Transcribe Portuguese Audio to Text — Free, No Upload

Turn Portuguese audio or video into text privately in your browser. 100% free, no per-minute limit, no account — nothing is ever uploaded.

Model

Spoken

Output

Timestamps [0:05]

⬇️ One-time download: 104 MB (Accurate model 73 MB + 31 MB audio engine, shared with the video tools) — only the first time: it stays saved in your browser, so next time it starts right away. (Translating or switching model needs its own file; a browser can clear this cache if storage runs very low.) Processing takes about 4-8 min per 10 minutes of audio on a typical laptop.

What exactly is downloaded — and is it safe?

• The speech model is OpenAI's Whisper, open source, converted to run in browsers. The audio engine is FFmpeg, the open-source tool behind most video software.
• They come from Hugging Face and unpkg — public repositories used by researchers and developers worldwide, not from us.
• They are not programs. Nothing is installed, no permission is asked, and your browser keeps them in its own cache exactly like it keeps a website's images.
• They can't read your files or touch anything outside this page. Remove them any time by clearing this site's data in your browser.
• Your audio never goes the other way: this site has no server receiving files.

Drop your video or audio here or click to choose

transcribed on your device by OpenAI's Whisper — your recording is not uploaded

ℹ️ Use a recording on your device that you own or have permission to process. This tool cannot fetch YouTube or social video links. Start with a short clip. Larger models can improve accuracy but require more memory and download data. Processing time varies by device, model and recording length. Always review the result.

Portuguese speakers generate more voice than text. Brazil runs on WhatsApp audio, Portugal's meetings and interviews are recorded on phones, and Portuguese lectures, podcasts and sermons pile up as hours of sound that nobody can search or quote. This page converts that sound into written Portuguese. It opens with Portuguese already set as the language, so a noisy recording is not misidentified as Spanish, and it accepts both Brazilian and European Portuguese, with their different accents and vocabulary, without asking you to choose a country.

The transcription is done by OpenAI's Whisper model running inside your browser, in Private mode by default: the audio engine and the speech model are downloaded the first time, cached when possible, and after that the recording never leaves your device. That matters for a voice note from a client, a doctor's instructions or a family conversation. Choose Fast for a quick draft, Accurate for cleaner text, or Best when you can spare the memory and the wait. The output comes with accents and cedillas in place, and you can copy it or download a TXT; subtitle exports exist for video.

Treat the result as a draft. Names, numbers, addresses and slang are where any transcriber slips, and a crackly voice note recorded in a car will have more errors than a clear interview. Read it through before forwarding.

More sizes & options

Frequently asked questions

Does it work with WhatsApp voice notes?: Yes. Export the message (long-press → Share) and drop the .opus file here — no conversion needed first.
Brazilian or European Portuguese?: Both. The model handles the two variants and their accents; you do not need to choose a country.
Do I have to pick the language?: No. On the main transcription page the spoken language is detected automatically. This page simply starts with Portuguese already selected, which saves a step and removes any doubt when the audio is noisy.
Which model should I pick for Portuguese?: Portuguese is well covered by Whisper, so Fast is fine for a clear voice note and Accurate for interviews and meetings. Switch to Best for noisy recordings, strong regional accents or when every word matters.
Are accents and punctuation included in the transcript?: Yes. The model writes standard Portuguese with accents, cedillas, tildes and punctuation, so the text can be pasted into a document without clean-up. Review proper names and numbers, which are the usual weak points.

Guides

Related tools

🔒 Private mode (the default): your recording never leaves your device. Instant mode is optional, you switch it on yourself, and it sends the audio to the browser's speech service.