domains / families / research / icassp-experiments

icassp-experiments

separate

Use when designing or auditing ICASSP experiments across signal-processing modalities — matching the metric to the task law (WER, SI-SDR, PESQ/STOI, EER/minDCF, PSNR/SSIM, BER, RMSE), anchoring baselines to current strong methods and standard corpora, sweeping the operating condition, and reporting spread over runs within the four-page limit.

When to use

• designing ICASSP experiments
• auditing ICASSP experiments
• matching metric to task
• anchoring baselines to current strong methods
• sweeping operating conditions
• reporting spread over runs
• ablation design
• experiment readiness assessment

When not to use

• topping a leaderboard by any margin
• using novel or convenient metrics
• using weak or dated references
• reporting single-condition numbers without sweep

Skill page

Source