Speech-to-Text via Crazyrouter
Transcribe audio files using OpenAI Whisper through Crazyrouter.
Script Directory
Agent Execution:
SKILL_DIR= this SKILL.md file's directory- Script path =
${SKILL_DIR}/scripts/main.mjs
Step 0: Check API Key ⛔ BLOCKING
[ -n "${CRAZYROUTER_API_KEY}" ] && echo "key_present" || echo "not_set"
| Result | Action |
|---|---|
key_present | Continue |
not_set | Ask user to set CRAZYROUTER_API_KEY. Get key at https://crazyrouter.com |
Usage
# Transcribe audio
node ${SKILL_DIR}/scripts/main.mjs --input audio.mp3
# Save to file
node ${SKILL_DIR}/scripts/main.mjs --input meeting.m4a --output transcript.txt
# Specify language hint
node ${SKILL_DIR}/scripts/main.mjs --input chinese.mp3 --language zh
# Translate to English
node ${SKILL_DIR}/scripts/main.mjs --input french.mp3 --translate
Options
| Option | Description | Default |
|---|---|---|
--input <file> | Audio file (required) | — |
--output <file> | Save transcript to file | stdout |
--language <code> | Language hint (ISO 639-1) | auto-detect |
--translate | Translate to English | false |
--model <id> | whisper-1 | whisper-1 |
Supported Formats
mp3, mp4, mpeg, mpga, m4a, wav, webm