getscript

A fast, Unix-friendly CLI for fetching transcripts from Apple Podcasts. You don't need to transcribe much of anything nowadays because Apple is transcribing everything for us. 'getscript' lets you use the transcripts in a more human / text friendly manner. The idea behind getscript is to put the transcripts to use in a meaningful way.

For example - let's say you're curious about AI Slop as a podcast topic:

$ getscript --search "AI Slop" --list <removing list of 10 podcasts and their podcast id's> $ getscript 1000730374732 | claude -p "Summarize the top 5 Points"

Of course you can also export to TTML, JSON, markdown and run from cron if you want to schedule things.

Install

pip install getscript

Or via Homebrew:

brew install outerbanks73/tap/getscript

Requires Python 3.10+. Apple Podcasts transcripts require macOS 15.5+ with Xcode CLI tools.

Quick Start

# Fetch from an episode URL
getscript "https://podcasts.apple.com/us/podcast/the-daily/id1200361736?i=1000753754819"

# Fetch from a bare episode ID
getscript 1000753754819

# Raw TTML XML output
getscript 1000753754819 --ttml

# Search Apple Podcasts interactively (requires fzf)
getscript --search "artificial intelligence"

Output Formats

Output formats are mutually exclusive (--json, --ttml, --markdown). The --timestamps flag can be combined with any format.

# JSON piped to jq
getscript EPISODE_ID --json | jq '.segments[].text'

# Markdown with timestamps
getscript EPISODE_ID --markdown --timestamps > notes.md

# Write to file
getscript EPISODE_ID -o transcript.txt

Search

# Search Apple Podcasts (requires fzf)
getscript --search "climate change"

# List results without fzf
getscript --search "topic" --list --limit 20

Stdin & Batch Processing

# Read episode ID from stdin
echo "1000753754819" | getscript -

# Batch process a list of IDs
cat ids.txt | xargs -n1 getscript --no-upload --quiet

# Disable upload via environment variable
GETSCRIPT_UPLOAD=0 getscript EPISODE_ID

# Silent file output for scripting
getscript EPISODE_ID --quiet --no-upload -o out.txt

Shared Transcript Library

Fetched transcripts are automatically submitted to voxlytranscribes.com for indexing and enrichment. To disable:

getscript EPISODE_ID --no-upload
GETSCRIPT_UPLOAD=0 getscript EPISODE_ID

Shell Completions

getscript --completions bash >> ~/.bashrc
getscript --completions zsh >> ~/.zshrc
getscript --completions fish > ~/.config/fish/completions/getscript.fish

Configuration

Config file: ~/.config/getscript/config.json

Environment variables:

Precedence: config file < environment variables < CLI flags.

See examples/config.example.md for a full annotated reference.

How It Works

Compiles a small Obj-C helper that calls Apple's private AMSMescal framework (FairPlay DRM) to generate a bearer token. That token authenticates requests to the AMP API, which returns TTML transcripts. The token is cached at ~/.cache/getscript/apple_token for 30 days. This is the only open-source tool that can fetch Apple Podcasts transcripts programmatically.

Shared Transcript Library

Every fetch contributes to the shared library at voxlytranscribes.com. Submissions go through a quarantine pipeline: server-side re-fetch, content hash verification, and provenance tracking before promotion to the canonical library. See the technical docs for details.

Exit Codes

Testing

87 tests across 9 modules. CI matrix: Python 3.10–3.13 on Ubuntu and macOS.

pytest -v

Contributing

See CONTRIBUTING.md.

Changelog

See CHANGELOG.md.

License

MIT