Latin America · Brazil
Portuguese (Brazilian) conversational speech dataset
2,000 hours of spontaneous Portuguese (Brazilian) conversations recorded in Brazil across 200 vetted native speakers. Dual-channel audio with time-aligned transcripts, speaker metadata (age, gender, region, device), and full chain-of-consent documentation. Licensed for commercial ASR, TTS, and voice AI training.
Dataset specifications
| Language | Portuguese (Brazilian) |
|---|---|
| Recorded in | Brazil |
| Speech type | Spontaneous two-person conversations |
| Volume | 2,000 hours |
| Speakers | 200 vetted native speakers |
| Audio | Dual-channel (one channel per speaker) |
| Transcripts | Time-aligned, human-verified |
| Metadata | Age, gender, region, recording device |
| License | Commercial (training & evaluation) |
| Price | $65 per hour of audio |
Need more hours, another dialect or your own recording setup? We collect speech data to spec.
Built for
Training and evaluating speech models that must handle real Portuguese (Brazilian) speech:
- Automatic speech recognition (ASR / STT)
- Text-to-speech & voice synthesis prosody
- Voice agents and IVR for Brazil
- Speaker diarization & turn-taking models
- Speech LLM / audio-understanding fine-tuning
Why conversational data?
Models trained on read speech consistently underperform on spontaneous audio. Disfluencies, overlap, and natural speech rates aren’t in the training distribution. This dataset closes that gap for Portuguese (Brazilian). Read more in our guide to conversational vs. scripted speech data.
Free samples
Audit the Portuguese (Brazilian) data yourself
Share your company email and we’ll email you representative audio, matching transcripts, and the metadata schema. Check audio quality, transcript accuracy, and speaker diversity before you spend a dollar.
Public alternatives
Public Brazilian Portuguese speech data vs. this dataset
| Source | What it is | License | Commercial training |
|---|---|---|---|
| CORAA ASR (opens in a new tab)NILC and the Tarsila project (C4AI, University of São Paulo) | Brazilian Portuguese ASR corpus from ALIP, C-ORAL Brasil, NURC-Recife, SP-2010 and TEDx; spontaneous, prepared and read speech. | CC BY-NC-ND 4.0 | Not allowed |
| Multilingual LibriSpeech (MLS) Portuguese (opens in a new tab)Meta AI (via OpenSLR) | Portuguese read audiobooks from LibriVox with transcripts. | CC BY 4.0 | Allowed |
| Mozilla Common Voice Portuguese (Scripted Speech) (opens in a new tab)Mozilla Common Voice (via Mozilla Data Collective) | Portuguese sentences read by volunteers, with a curated Brazilian Portuguese subset; no Spontaneous Speech release for Portuguese. | CC0 1.0 | Allowed |
| Google FLEURS (pt_br) (opens in a new tab)Google | Read sentences from the FLoRes benchmark, locale pt_br. | CC BY 4.0 | Allowed |
CORAA has spontaneous Brazilian Portuguese but a non-commercial, no-derivatives license; this dataset is two-person conversation recorded in Brazil with dual-channel audio and a commercial license.
Licenses as published by each source, checked October 2026. Check the current terms before use.
FAQ
Frequently asked questions
What does the Portuguese (Brazilian) speech dataset include?
2,000 hours of spontaneous two-person conversations from 200 vetted native speakers recorded in Brazil, delivered as dual-channel audio with time-aligned transcripts, speaker metadata (age, gender, region, device), and consent documentation.
What makes Brazilian Portuguese hard for speech recognition?
Brazilian Portuguese keeps its vowels open and full, and in much of the country "t" and "d" before an "i" sound come out as "ch" and "j". Everyday speech uses "você" and "a gente", and accents from São Paulo, Rio and the Northeast differ noticeably, so speaker coverage matters.
Is there free Brazilian Portuguese speech data?
Yes, Multilingual LibriSpeech Portuguese, Mozilla Common Voice Portuguese and Google FLEURS are free for commercial use. CORAA ASR is non-commercial only. This dataset is spontaneous two-person conversation recorded in Brazil, with speaker metadata, one channel per speaker and signed commercial-use consent.
Can I use this Portuguese (Brazilian) dataset commercially?
Yes. The dataset is licensed for commercial use, including training ASR, TTS, and other voice-AI models. Every speaker has signed explicit commercial-use consent, and we provide the full consent chain for your legal review.
How much does the Portuguese (Brazilian) speech dataset cost?
Pricing is $65 per hour of audio. Volume discounts are available for full-dataset and multi-language purchases. Contact us for a quote.
Can I evaluate samples before buying?
Yes. Request samples with your company email, and we'll email you representative audio files and matching transcripts so your team can audit quality before committing.
Related datasets
You might also need
$85/hr