The researchers recruited dialect-community coordinators, who selected culturally relevant images across 10 topics.
New benchmark measures speech recognition across 17 Arabic dialects
Almieyar uses newly recorded, community-curated speech and finds that the best system still errs on roughly one word in three.
Independent
Omid Ghahroodi · Anas Madkoor · Dima Faris Al Saudi · Fagr Tahir · Malak Annan · Talha shahid javad allah rakha · +34 more
Research Digest··2 min read
The authors introduce Almieyar, a zero-shot speech recognition benchmark covering 17 Arabic dialects from six dialect families, with all audio newly recorded for the task.
Why this paper
Independent
In one line
GPT-4o-transcribe achieves 35% WER on 17 Arabic dialects, but performance varies considerably across groups.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ·No stated limitations found
- ✓Reports numbers on named benchmarks
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
Concepts
§