BIPA is a grapheme-to-phoneme dataset for Brazilian Portuguese, created by thiagomonteles. It provides transcriptions in the International Phonetic Alphabet and includes dialectal labels, derived from Wiktionary under CC BY-SA 4.0. The dataset features consistent symbol normalization and standardization of dialectal labels.
Use Cases
- Train grapheme-to-phoneme conversion models based on IPA transcriptions.
- Analyze dialectal variations in Brazilian Portuguese pronunciation based on dialectal labels.
- Benchmark phonetic transcription systems based on the standardized IPA inventory.
- Study the relationship between orthography and phonology in Brazilian Portuguese.
Strengths
- Includes a normalized inventory of 107 segmental letters and 44 diacritics.
- Provides standardized dialectal labels for phonetic variation analysis.
- Derived from a known source (Wiktionary) under a clear license (CC BY-SA 4.0).
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- Last updated 2026-04-30 18:31:37; freshness should be verified.
Provenance
- Source
- Wiktionary
- Collection Method
- Derived and processed from Wiktionary entries.
- Geography
- Brazilian Portuguese dialects