Turn Portuguese meetings, interviews and lectures into searchable text with speaker labels — Brazilian, European and African Portuguese.
iOS and Android. 30 minutes free, no card.
Does SozAI transcribe Portuguese?
SozAI transcribes Portuguese audio and video into text in Brazilian, European and African Portuguese. It labels each speaker automatically, timestamps every word, writes the full orthography including nasal vowels and cedillas, and exports to TXT, DOCX, PDF, SRT and VTT. Recordings are processed in EU data centers and encrypted at rest, and every account gets 30 minutes a month without a card.
Português, written the way it is actually written.
Specification
| Native name | Português |
|---|---|
| ISO 639-1 code | pt |
| Writing system | Latin |
| Text direction | Left to right |
| Speakers worldwide | ~260 million |
| Varieties covered | European, Brazilian, African |
| Speaker labels | Yes — automatic diarization |
| Timestamps | Word level |
| Summaries | Yes, in the same language |
| Export formats | TXT, DOCX, PDF, SRT, VTT |
| Maximum file size | 500 MB |
| Free tier | 30 minutes per month, no card |
| Where data is stored | EU data centers, AES-256 at rest |
| Public transcripts in this language | 395 |
Workflow
Record straight into the app, upload a file up to 500 MB, or paste a YouTube link. SozAI reads m4a, mp3, wav, aac, mp4 and mov.
Choose Brazilian or European Portuguese when you know which one you recorded, or leave detection on. The two are treated as separate targets because they differ far more in speech than in writing.
The transcript arrives with speaker labels and word timestamps. Search it, ask it questions, generate a summary in Portuguese, or export SRT and VTT subtitles.
Field notes
They share an orthography and diverge sharply in the audio. European Portuguese reduces unstressed vowels so heavily that whole syllables become inaudible to a listener trained on Brazilian speech, while Brazilian Portuguese keeps them open and adds a palatal "ti/di". A model tuned on one performs noticeably worse on the other, which is why the variety is a setting and not a detail.
The difference between "pau" and "pão", or "vi" and "vim", lives in nasalisation rather than in a consonant. Compressed phone audio is exactly where nasality degrades first. Transcripts that read almost right but flip a nasal vowel are worse than obvious errors, because nothing looks wrong on the page.
Brazil largely uses "você" with third-person verb forms; Portugal and southern Brazil use "tu" with second-person forms, and formal registers add "o senhor". These choices ripple through every verb in the sentence, so a transcript that normalises them silently rewrites the tone of the conversation.
Business recordings are full of terms that never appear in general-purpose training data: local product names, gíria, and Portuguese-English hybrids common in Brazilian tech teams. Word-level timestamps matter here, because the fastest way to fix an unfamiliar term is to jump straight to the second it was said.
From the archive
395 public Portuguese transcripts so far — every one produced by the same engine described above. Read them in full, no account needed.
Ribeirão do Tempo Capitulo 1 — Transcript
Mal-estar, Sofrimento e Sintoma | Christian Dunker — Transcript
MD5 – Processos Operacionais | CEO — Transcript
A FORMA MAIS EFICIENTE DE GANHAR DINHEIRO COM A INTERNET — Transcript
YouTube Video — Transcript
MD5 – Gestão Empresarial | Financeira — TranscriptSozAI transcribes over 100 languages. These have a page of their own.
Answers
Short answers you can quote. Longer ones live in the sections above.
Yes, and it treats them as distinct targets rather than one language with an accent. You can set the variety yourself or leave automatic detection on. If you record a call with participants from both, the transcript stays in one file — the setting affects how the audio is interpreted, not how the file is split.
Yes. The transcript is written in full Portuguese orthography: ã, õ, ç, and the acute, grave and circumflex accents. This is not cosmetic — dropping a tilde turns one word into another, and a transcript without diacritics cannot be pasted into a document or a subtitle file without manual repair.
Yes. Diarization works on the audio, so it separates voices in Portuguese exactly as it does in English. Each turn is labelled and timestamped, which is what makes a recorded interview or a deposition usable as evidence rather than a wall of text.
SozAI exports SRT and VTT with word-level timing and UTF-8 text, so accented characters survive the round trip into YouTube, Premiere Pro, Final Cut Pro and DaVinci Resolve. You get subtitle files that need timing tweaks at most, not a character-encoding rescue.
Every account gets 30 minutes per month free, with no card. Beyond that SozAI is a flat monthly subscription with a shared pool of minutes; the language you transcribe does not change the price. Current plans are listed on the pricing page.
Where to use it
Everything below works in Portuguese, with the same speaker labels and word timestamps.
Transcribe your first Portuguese recording
Record, upload a file or paste a link. Speakers, timestamps and a summary come back together.
Free on iOS and Android. No account required for YouTube transcripts.
Point your camera
Scan the code to open the download page on iOS or Android. 30 minutes of transcription free, no card.
