Skip to main content

Documentation Index

Fetch the complete documentation index at: https://notewise.click/docs/llms.txt

Use this file to discover all available pages before exploring further.

notewise process accepts a YouTube video URL or a bare 11-character video ID. It fetches metadata and transcript text, generates notes with the configured LiteLLM model, writes files locally, and stores run metadata in SQLite.

Supported video inputs

InputExample
Watch URLhttps://www.youtube.com/watch?v=VIDEO_ID
Short URLhttps://youtu.be/VIDEO_ID
Embed URLhttps://www.youtube.com/embed/VIDEO_ID
/v/ URLhttps://www.youtube.com/v/VIDEO_ID
Shorts URLhttps://www.youtube.com/shorts/VIDEO_ID
Bare video IDVIDEO_ID

Basic command

notewise process "https://youtu.be/VIDEO_ID"

Common options

OptionDefaultUse it for
--model, -mgemini/gemini-2.5-flashOverride the LiteLLM model for one run.
--output, -o./outputWrite files to a specific folder.
--formatmdWrite md, html, pdf, docx, or a comma-separated list.
--language, -lenSpecifies the transcript language; this option can be repeated.
--target-languageEnglishGenerate notes in another language.
--temperature, -t0.7Change LLM temperature, range 0.0 to 1.0.
--max-tokens, -kprovider defaultCap LLM output tokens.
--force, -FfalseIgnore cache and regenerate.
--no-uifalseDisable the Rich live dashboard.
--verbose, -vfalseWrite DEBUG diagnostics to the session log.
--quizfalseGenerate a quiz Markdown file.
--export-transcriptnoneExport transcript as txt or json.
--timestampsfalsePrefix generated chapter headers with start times.
--chapter-directory-outputfalseWrite per-chapter Markdown files when YouTube chapters exist.
--cookie-file, --cookiesYOUTUBE_COOKIE_FILERead restricted videos with a Netscape cookies file.
--use-combine-chunkfalseUse deprecated legacy chunk-combine behavior.
--use-combine-chunk is legacy behavior. The default long-transcript path uses boundary stitching.

Option examples

notewise process "https://youtu.be/VIDEO_ID" --format md,docx
notewise process "https://youtu.be/VIDEO_ID" --quiz --export-transcript json
notewise process "https://youtu.be/VIDEO_ID" --force --no-ui
Chapter-aware generation is automatic when YouTube chapters exist. By default, the final notes are bundled into one output file. Use --chapter-directory-output only when you want separate per-chapter Markdown files.
Last modified on May 1, 2026