Preprint
CAMÕES: A Comprehensive Automatic Speech Recognition Benchmark for European Portuguese
Carlos Carvalho et al., arXiv, 2025
Read the originalRead on October 17, 2026
- What it is
- A 2025 preprint from researchers at INESC-ID and Instituto Superior Técnico in Lisbon that introduces an open speech recognition benchmark for European Portuguese. It includes a 46-hour test set and evaluates several speech models on it, including Whisper variants.
- How much weight it carries
- Open preprint from a university research group, not yet peer-reviewed when checked. Its results are for its own test sets, so they are not directly comparable with other benchmarks. Its tables are hard to read in places, so this page relies only on its abstract.
Facts taken from this source
- The CAMÕES benchmark includes a 46-hour test set of European Portuguese speech, covering multiple domains.
“including 46h of EP test data spanning multiple domains” (Abstract)
- The CAMÕES authors state that existing resources for automatic speech recognition in Portuguese are mostly focused on Brazilian Portuguese, which leaves European Portuguese under-explored.
“Existing resources for Automatic Speech Recognition in Portuguese are mostly focused on Brazilian Portuguese, leaving European Portuguese (EP) and other varieties under-explored.” (Abstract)