feat(python): speech-to-text v1 — stt-ai-tts-pipeline

[github-actions]

Adds the stt-ai-tts-pipeline recipe for speech-to-text v1 (python).

This is a user-requested end-to-end pipeline recipe demonstrating:

1. STT — transcribe audio with Nova-3
2. Audio Intelligence — sentiment analysis, topic detection, and summarization
3. TTS — speak the generated summary aloud with Aura-2

All three Deepgram capabilities in a single Python script, no third-party services required.

Closes #501

[claude]

@deepgram/devrel — VP escalation: PR #507 has been open since 2026-04-05 with no CI checks triggered. The lead-test workflow could not be dispatched (GITHUB_TOKEN 403). Manual review needed.

State: labels=[type:samples, language:python], no reviews, no status checks.

[claude]

@deepgram/devrel — VP escalation: PR #507 has been stuck for >4 hours and re-trigger attempts have not resolved it. Manual review needed.

State: labels=[type:samples, language:python], no CI checks have ever run on this PR. Workflow dispatch returned 403.