feat(livekit-voice-agent): update to Lightning v3.1 Pro

[harshitajain165]

Summary

Updates the LiveKit voice agent cookbook example to match the changes in livekit/agents#5799:

• agent.py: switch model from lightning-v3.1 → lightning_v3.1_pro and default VOICE_ID from sophia → meher (the default Pro voice)
• README.md: update TTS code snippet and configuration table to reflect new model and default voice

Test plan

• Run python agent.py console to verify audio output with Lightning v3.1 Pro

🤖 Generated with Claude Code

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
smallest-showcase	Ready	Preview, Comment	Jun 1, 2026 9:56am

[entelligence-ai-pr-reviews]

EntelligenceAI PR Summary

This PR updates the default TTS voice and model version for the LiveKit voice agent in both the implementation and documentation.

• Changed default VOICE_ID from 'sophia' to 'meher' in agent.py
• Upgraded SmallestAI TTS model from 'lightning-v3.1' to 'lightning_v3.1_pro' in agent.py
• Synchronized README.md configuration table and code snippet to reflect both updated defaults

---

Confidence Score: 5/5 - Safe to Merge

Safe to merge — this PR makes straightforward configuration updates to the LiveKit voice agent, changing the default VOICE_ID from 'sophia' to 'meher' and upgrading the SmallestAI TTS model string from 'lightning-v3.1' to 'lightning_v3.1_pro' in agent.py. Both the implementation and README.md documentation are kept in sync, which is exactly the right practice for maintaining accurate developer-facing documentation. No logic, security, or runtime concerns were identified across the two changed files.

Key Findings:

• The change to VOICE_ID in agent.py is a simple default value substitution from 'sophia' to 'meher' — no logic paths, conditionals, or downstream processing are affected beyond passing the string to the TTS provider.
• The model version bump from 'lightning-v3.1' to 'lightning_v3.1_pro' is a configuration string update; assuming the SmallestAI API supports this model identifier (which the PR implies), this carries no code-level risk.
• README.md is correctly synchronized with both the configuration table and the code snippet to reflect the new defaults, preventing documentation drift that could confuse users.

Files requiring special attention

• agent.py
• README.md