Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,341 papers · 148 categories

Trend · papers per month

0111 · Jan 201619922001200920182026
1 result for re-speaking

This paper compares automatic metrics for re-speaking quality assessment.

problem Estimating the quality of re-speaking results is challenging.
method Comparing and adapting automatic evaluation metrics (BLEU, EBLEU, NIST, METEOR, etc.) to re-speaking quality.
result Automatic metrics are compared to human-derived NER metric for re-speaking quality.