ADLIB DevTerm: Japanese Software Development Terminology Audio Test Set

A test audio dataset for the ADLIB language-aware ASR benchmark framework for Japanese. It contains 247 test cases with audio from 3 speakers, focusing on the DevTerm (software development terminology) domain. Reference transcripts and term annotations are provided in a separate JSONL file within the project's GitHub repository.

Use Cases

• Benchmarking ASR model accuracy on Japanese technical terminology based on the described domain focus.
• Evaluating speech recognition performance across different speakers based on the mention of 3 speakers.
• Testing language-aware ASR frameworks on domain-specific audio as part of the ADLIB benchmark.

Strengths

• Contains 247 test cases, providing a defined evaluation scale.
• Includes audio from 3 speakers, allowing for speaker variation analysis.
• Focuses on a specific domain (software development terminology), which likely increases relevance for technical ASR tasks.

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

GitHub repository 'holotherapper/adlib'.

Collection Method

Likely created as a test set for the ADLIB benchmark framework.

Freshness

Last updated 2026-04-13 13:55:24; freshness should be verified.