Earnings-22
Emerging5papers using it
2022first seen
The 'Earnings-22' dataset is a benchmark used to evaluate the performance of speech recognition and translation models, specifically focusing on punctuation and capitalization accuracy.
Papers using Earnings-22 (5)
- TRADE: Transducer-Augmented Decoder for Speech LLMEarnings-22: A Practical Benchmark for Accents in the WildInvestigating End-to-End ASR Architectures for Long Form Audio
TranscriptionHow Much Context Does My Attention-Based ASR System Need?Longer is (Not Necessarily) Stronger: Punctuated Long-Sequence Training
for Enhanced Speech Recognition and Translation