SEP-28-k
Emerging4papers using it
2022first seen
The 'SEP-28-k' dataset/benchmark contains stuttered speech samples and is used to evaluate the classification of disfluency types in speech processing research.
Papers using SEP-28-k (4)
- StutterZero and StutterFormer: End-to-End Speech Conversion for Stuttering Transcription and CorrectionOptimizing Multi-Stuttered Speech Classification: Leveraging Whisper's Encoder for Efficient Parameter Reduction in Automated AssessmentIntroducing ECAPA-TDNN and Wav2Vec2.0 Embeddings to Stuttering DetectionWhisper in Focus: Enhancing Stuttered Speech Classification with Encoder Layer Optimization