GCP speech result subcommand | Generated by AI

Home 2026.07

All changes are in place. Here’s a summary of what was done:

Changes made:

  1. ww/gcp_speech/result.py (new file) — the ww gcp-speech result <job-id> command:
    • Looks up the job ID in ~/.ww_gcp_jobs.json (the local jobs registry)
    • Checks GCS for completed transcription JSON files
    • If results exist: downloads them, prints the transcript, and saves a .md file alongside the original audio
    • If not yet complete: prints the console URL and suggests --wait to poll
    • Supports --wait flag for blocking mode (polls every 10s)
  2. ww/gcp_speech/transcribe.py — modified to:
    • Print a clear Job ID: line in the output (the basename/filename-stem)
    • Save job info to ~/.ww_gcp_jobs.json (operation name, GCS folder, audio file, language)
    • Direct users to ww gcp-speech result <job-id> --wait instead of gsutil cp

  3. ww/main.py — added result subcommand routing and help text

  4. ~/.zsh/completions/_ww — added result to gcp_speech_cmds for tab completion

Usage flow:

ww gcp-speech transcribe ~/Downloads/recording.mp3
  → Prints: Job ID: recording
  → Saves job to ~/.ww_gcp_jobs.json

ww gcp-speech result recording
  → Checks GCS for results, prints transcript if ready

ww gcp-speech result recording --wait
  → Polls until complete, then prints transcript

Generated by AI. Curating and sharing still takes effort. If you find it useful, feel free to donate. WeChat: @lzwjavaWeChat QR · X: @lzwjava · Say hi 👋

Back Donate