Spellcheck can't fix "chai GPT": building an LLM caption checker on $5 a month

说明与用途

Misheard 利用 LLM 检测并建议修正 YouTube 自动字幕中的误听词，精度达 78%，且费用控制在每月 5 美元以内。

问题

YouTube auto‑captions frequently mishear words, producing real‑word errors that standard spellcheckers cannot detect.

结果

Misheard improves real‑word error detection from 6/79 to 22/79 on hand‑checked videos, with 0.78 precision, and operates within a $5/month budget.

工作流

• Upload SRT/VTT file
• Local detectors flag non‑word and phonetic errors
• Chunk transcript (~400 words) and send to LLM with hints
• LLM returns misheard words and suggested fixes
• User reviews and accepts/rejects suggestions
• Export corrected transcript

证据

Over half were real‑word errors (55), and only 41 were non‑words.

来源

Read-through: The transcript goes out in chunks of about 400 words. Each chunk carries the local detectors' flags as hints, plus priming terms.

来源

That’s 3.7x and 8.3x the real‑word catches, with better precision, at about $0.05 per hour of audio.

来源