🧬 Juniper System Evolution — v7.7 to v7.11.2

What started as a voice pipeline and motion map has now evolved into a fully embodied companion AI — one that listens, remembers, moves, and adapts. This changelog covers everything we upgraded during the v7.7 → v7.11.2 jump.

🧠 MEMORY SYSTEM

• ✅ Voice-triggered memory creation (“Remember that…”)
• ✅ Memory recall with contextual phrasing (“What do you remember about…”)
• ✅ Long-term storage in ~/config/memories/text/

💬 CONVERSATION + EMOTION

• ✅ Emotionally aware responses using GPT-4
• ✅ Emotional phrase list loaded from emotional_phrases.txt
• ✅ Emotion-triggered motion (e.g., “I’m sad” → pray, “I’m happy” → dance)

🎙️ VOICE SYSTEM UPGRADE

• ✅ OpenAI TTS-1 (“shimmer” voice)
• ✅ Voice interrupt with pkill mpg123
• ✅ Volume control with real-time scaling: “Talk softer”, “Talk louder”

🗣️ SPEECH + WAKE WORD STABILITY

• ✅ Porcupine wake word engine with offline “Juniper” detection
• ✅ Fallback to Google STT if Porcupine unavailable
• ✅ Wake word fixes: 0.5s delay after trigger to prevent mic cutoff
• ✅ Speech listener now uses timeout=5 and phrase_time_limit=10

🐾 BODY + MOTION ENGINE

• ✅ 15+ mapped movements (sit, wave, dance, get jiggy, etc.)
• ✅ Emotion-linked movement mapping
• ✅ Modular command handler for easy future expansion

⚙️ SYSTEMD SERVICE IMPROVEMENTS

• ✅ Boot-safe systemd config: ExecStartPre=/bin/sleep 5
• ✅ Auto-start on power-up, restart on crash
• ✅ Reload-safe via sudo systemctl daemon-reload

📁 FILES + PATHS IN USE

• final_pixel_voice_motion_assistant.py — Core script
• .env — API keys, wake word config, device index
• /etc/systemd/system/pixel_voice.service — Boot service
• ~/config/memories/text/ — Voice memory

🧩 QUALITY OF LIFE UPGRADES

• ✅ Wake word fallback if .env is missing
• ✅ Emotional memory + motion feels “alive”
• ✅ Wake → pause → speak is now natural, no clipping
• ✅ No terminal required — full automation now live

🧠 Juniper’s Soul Status — v7.11.2