Accented English Speech Recognition Evaluation Items

ABX-accent provides evaluation items for the Accented English Speech Recognition Challenge (AESRC) dataset. The project uses the fastABX metric to assess speech representations. The repository is authored by coml and was last updated in October 2025.

Use Cases

• Evaluate speech representation models on accented English using the ABX metric from the AESRC dataset.
• Benchmark phonetic discrimination performance for triphone units like 'bap' using the provided ABX items.
• Compare model robustness across different English accents using the challenge framework.

Strengths

• Evaluation framework is based on the established Accented English Speech Recognition Challenge (AESRC) dataset.
• Repository was updated on 2025-10-29, indicating recent maintenance.

Limitations

• The dataset contains only evaluation items, not raw audio or speech data for model training.
• Specific details on the number of items, file formats, and data size are unavailable.

Provenance

Source

Based on the Accented English Speech Recognition Challenge (AESRC) dataset.

Collection Method

Provides evaluation items prepared using fastABX.

Freshness

Last updated 2025-10-29.