DirectSpeech2LLM: A Simple End-to-End Framework to Mitigate Prompt Overfitting in Speech-LLMs
DirectSpeech2LLM cuts speech-LLM prompt overfitting, generalizing from ASR-only training to translation and emotion tasks.
Stories per day · 30 days
4 stories, newest first · one card per story
Stories tagged #ctc
Pruned CTC for Memory-Efficient Large-Vocabulary ASR Training
Pruned CTC restricts alignment computation to target tokens and blank, cutting ASR training memory 5.1x for 180K-vocabulary LLMs with minimal accuracy loss.
A Training Criterion with Token-Level Tolerance to Transcription Ambiguity for Automatic Speech Recognition
Token-level wildcard paths in CTC cut ASR word error by tolerating localized transcript ambiguity.
Design of the IBM Granite 5.0 TurboCTC ASR Model
IBM details Granite 5.0 TurboCTC, a 470M-parameter Conformer-based English ASR model on the Open ASR speed-accuracy Pareto frontier, released on Hugging Face.