Buckeye

The 'Buckeye' dataset is a speech corpus that contains recordings of conversational American English, used to evaluate the performance of speech alignment systems.

Papers using Buckeye (8)

Gradient-Based Speech-to-Text Alignment for Any ASR Model: From CTC to Speech LLMs2026Multilingual Word-Level Forced Alignment with Self-Supervised Representations and Learned Dynamic Programming2026BFA: Real-time Multilingual Text-to-speech Forced Alignment2025Back To Supervision: Boosting Word Boundary Detection Through Frame Classification2024Tradition or Innovation: A Comparison of Modern ASR Methods for Forced Alignment2024 · 1 citesUnsupervised Speech Segmentation and Variable Rate Representation Learning using Segmental Contrastive Predictive Coding2021Phoneme Segmentation Using Self-Supervised Speech Models2022Back to Supervision: Boosting Word Boundary Detection through Frame Classification2024