WhisTLE: Deeply Supervised, Text-Only Domain Adaptation for Pretrained Speech Recognition Transformers 文章

ArXiv CS.CL2026-05-26NEWSen作者: Akshat Pandey, Karun Kumar, Raphael Tang
查看原文 →
关系图谱

WhisTLE: Deeply Supervised, Text-Only Domain Adaptation for Pretrained Speech Recognition Transformers · 相关技术

相关技术

TTSVAEencoder-decoderASRencoder-decoder ASR modelstext-to-speech (TTS)variational autoencoder (VAE)