Fisher dataset
Emerging9papers using it
2021first seen
The Fisher dataset is a collection of 2000 hours of two-channel raw conversational audio used to evaluate models for generating naturalistic spoken dialogues without text or labels.
Papers using Fisher dataset (9)
- LESS: Large Language Model Enhanced Semi-Supervised Learning for Speech Foundational Models Using in-the-wild DataSLIDE: Integrating Speech Language Model with LLM for Spontaneous Spoken Dialogue GenerationSEAL: Speaker Error Correction using Acoustic-conditioned Large Language ModelsGenerative Spoken Dialogue Language ModelingComparison of SVD and factorized TDNN approaches for speech to textLexical Speaker Error Correction: Leveraging Language Models for Speaker Diarization Error CorrectionTowards Real-World Streaming Speech Translation for Code-Switched SpeechAG-LSEC: Audio Grounded Lexical Speaker Error CorrectionImproving Speech Recognition Error Prediction for Modern and Off-the-shelf Speech Recognizers