QASR: QCRI Aljazeera Speech Resource with 2,000 Hours

2,000 hours of transcribed Arabic speech collected from Aljazeera News Channel broadcasts. QASR is a large-scale corpus created by QCRI, featuring multi-layer annotation and covering multiple Arabic dialects and code-switching speech. The dataset was last updated on the platform in October 2025.

Use Cases

• Train automatic speech recognition models based on the large volume of transcribed Arabic speech.
• Study dialectal variation in Arabic based on the coverage of multiple dialects mentioned in the description.
• Develop models for code-switching speech recognition based on the inclusion of code-switching data.
• Benchmark speech processing systems on a large-scale, lightly supervised corpus.

Strengths

• Large scale with approximately 2,000 hours of speech data.
• Features multi-layer annotation and linguistic segmentation.
• Covers multiple Arabic dialects and includes code-switching speech.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file formats are unknown, which may limit suitability assessment.

Provenance

Source

QCRI (Qatar Computing Research Institute).

Collection Method

Collected from Aljazeera News Channel broadcasts.

Time Range

null

Freshness

Last updated 2025-10-13 10:14:57; freshness should be verified.

Geography

null