MMAE: A Massive Multitask Audio Editing Benchmark

MMAE is a benchmark for instruction-based audio editing, serving as a comprehensive evaluation testbed. It covers 7 distinct audio modalities, including sound, speech, and music. The dataset was created by BoJack and was last updated on HuggingFace in June 2026.

Use Cases

• Benchmarking audio editing models based on the described multitask structure.
• Training instruction-following AI for audio manipulation across 7 modalities.
• Evaluating model performance in real-world audio editing scenarios mentioned in the overview.

Strengths

• Designed as the first comprehensive evaluation testbed for general-purpose instruction-based audio editing.
• Extends to a broad spectrum of real-world scenarios, encompassing 7 distinct audio modalities.

Limitations

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

Provenance

Source

HuggingFace dataset uploaded by author 'BoJack'.

Freshness

Last updated 2026-06-05 13:02:34; freshness should be verified.