My Weird Prompts
Teaching Speech Models to Transcribe Bilingual Slang
22:56
Fine-tuning a speech recognition model like Parakeet on English-Hebrew code-switching reveals sharp diminishing returns on recorded audio. The earliest seventy-seven minutes of speech nearly halved the word error rate, while subsequent hours provided smaller, more expensive gains. The distinction between word error rate and specific term recall proves vital, especially when teaching consistent…