Episode Details

Teaching an automated speech recognition model new vocabulary is a closed-set problem, but teaching it formatting behaviors is an open-ended mapping rule. Fine-tuning a non-autoregressive FastConformer to convert individually spelled letters into unified words requires learning conditional structures rather than just memorizing vocabulary. A look at inverse text normalization, contextual biasing limits, and grammar transducers reveals why behavioral training requires distinct design choices over raw volume. Key sources: • NVIDIA, model card, created 2025-08-04, last modified 2026-08-05 — https://huggingface.co/nvidia/parakeet-tdt-0.6b-v3 • Myrtle.ai, CAIMAN-ASR conditional training docs — https://caiman-asr.myrtle.ai/training/conditional_training.html • OpenAI cookbook, archived — https://developers.openai.com/cookbook/examples/whisper_correct_misspelling Episode 6444 — myweirdprompts.com/n/6444 (permanent short link: myweirdprompts.com/011153) Listen on Android: MWP FM Player — myweirdprompts.com/apk