PF-STAR
Emerging5papers using it
2022first seen
The 'PF-STAR' dataset/benchmark is used to evaluate speech recognition systems specifically for children's speech, focusing on the challenges posed by variations in speaker types and pitches.
Papers using PF-STAR (5)
- Zero-Shot KWS for Children's Speech using Layer-Wise Features from SSL ModelsCan Layer-wise SSL Features Improve Zero-Shot ASR Performance for Children's Speech?Layer-Wise Analysis of Self-Supervised Representations for Age and Gender Classification in Children's SpeechA Wav2vec2-Based Experimental Study on Self-Supervised Learning Methods to Improve Child Speech RecognitionFilter-based Discriminative Autoencoders for Children Speech Recognition