Ibibio and Efik Speech Corpus Sample from Akwa Ibom, Nigeria

A 1-hour sample of labeled speech data for the Ibibio and Efik languages. The data was collected from native speakers in Akwa Ibom, Nigeria by Scuba, a voice data infrastructure company. The sample was posted on Hugging Face on May 19, 2026.

Use Cases

• Train speech recognition models based on labeled Ibibio and Efik audio.
• Evaluate model performance on low-resource African languages based on the sample data.
• Benchmark speech synthesis systems for Ibibio and Efik based on native speaker recordings.
• Develop language identification tools for Nigerian languages based on the speech corpus.

Strengths

• Contains 1 hour of labeled speech data.
• Data is collected from native speakers in a specific region (Akwa Ibom, Nigeria).

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Scuba (scubavoice)

Collection Method

Collected and labeled by a voice data infrastructure company.

Freshness

Last updated 2026-05-19 05:55:19; freshness should be verified.

Geography

Akwa Ibom, Nigeria