CV-18 NER: Arabic Speech Named Entity Recognition Benchmark

CV-18 NER is the first publicly available dataset for Named Entity Recognition from Arabic speech. It was created by augmenting the Arabic Common Voice 18 corpus with manual NER annotations following the fine-grained Wojood schema, which covers 21 entity types. The dataset was authored by Elyadata and last updated on 2026-05-19.

Use Cases

• Benchmarking pipeline systems (ASR + text NER) based on the described Arabic speech corpus.
• Evaluating end-to-end speech NER models as mentioned in the description.
• Researching NER in low-resource settings using the provided benchmark.
• Studying entity recognition for morphologically complex languages based on the Arabic and Wojood schema context.

Strengths

• First publicly available dataset for Named Entity Recognition from Arabic speech.
• Annotations follow the fine-grained Wojood schema covering 21 entity types.
• Provides a benchmark for both pipeline and end-to-end model evaluation.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and dataset size are unknown, which may limit suitability assessment.
• Freshness should be verified as the last update timestamp is 2026-05-19.

Provenance

Source

Elyadata via Hugging Face.

Collection Method

Created by augmenting the Arabic Common Voice 18 corpus with manual NER annotations.

Freshness

Last updated 2026-05-19 08:28:32.