Commit Graph
4 Commits
Author SHA1 Message Date
jpmschweitzerandClaude Opus 4.6 9f34d030d7 feat(voice): complete Spike 2 voice pipeline with quality-tested prompt engine
Spike 2 delivers the full voice pipeline: queue → worker pool → sr-voice
child process (stdio JSONL) → cache → disk. Three rounds of quality testing
with Paula, Mellanie, and Gestalt produced iterative prompt improvements.

Prompt engine (prompt_builder.rs):
- Example-based epistemic marker integration (not keyword lists)
- Length-aware Angry tell variant (preserves facts on long content)
- Double-prompt technique: REMEMBER block repeats constraints near OUTPUT:
- Imperative injection framing (composition engine controls frequency)
- Anti-invention constraint ("do not add information not in the input")
- Universal RULES cleaned: worldbuilding moved to culture personas

Worker pool (worker.rs):
- Output post-processor strips after first newline (prevents prompt leakage)
- Watchdog poll loop (1s ticks) replaces blocking sleep for cancel
- Child health check before writing (try_wait)

Test infrastructure:
- voice_pipeline.rs: end-to-end test, auto-detects real sr-voice or mock
- voice_quality_batch.rs: 39 edge-case prompts for quality review
- mock-stdio.sh: Python JSONL mock for CI (no model needed)
- Makefile targets: test-voice-mock, test-voice-real

Quality results (Gemma 2B Q4_K_M, CPU ~13 t/s):
- Epistemic markers: naturally integrated (round 1 comma-lists fixed)
- Tell differentiation: 3/5 working (Nervous, Guarded, Angry)
- Information preservation: ~90% (up from ~70%)
- Prompt leakage: eliminated
- Open: Friendly/RoutineDeviation tells inert (#651), Factual bypass (#650)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-07 19:20:06 +01:00
jpmschweitzerandClaude Opus 4.6 e93a9e8b70 fix(voice): address PR review findings — 3 critical, 5 warning, 4 suggestion
Critical fixes:
- Pause mechanism: workers now hold requests during pause instead of
  dropping them. Queue and worker pool share the same AtomicBool flag
  via VoiceQueue::paused_flag(). Submit() rejects while paused.
- Seed type: sr-voice accepts u64 seeds over IPC (explicit u32 truncation
  for llama.cpp sampler, documented).

Warning fixes:
- HashMap → BTreeMap in cache.rs and worker.rs (D-010 determinism mandate).
  Added Ord derives to CacheKey, ContentType, TellCategory.
- VoicePipe::generate() watchdog kills child after 120s timeout to prevent
  indefinite blocking on read_line.
- VoiceCacheStore Drop impl calls save_all() on shutdown.
- trait-modifiers.ron: fixed 3 wrong trait names (Impulsive→Compassionate,
  Methodical→Incurious, Stubborn→Ruthless) to match PersonalityTrait enum.

Suggestion fixes:
- Worker spawn: log error + reduce pool instead of panic on thread failure.
- on_battery(): added macOS detection via pmset.
- Epistemic markers: lowercased constants, removed redundant to_lowercase().
- cache.rs: documented non-atomic write tradeoff.
- queue.rs: reprioritize() bypasses pause check (it runs during pause).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-07 18:11:46 +01:00
jpmschweitzerandClaude Opus 4.6 0b0fa8c04e refactor(voice): replace HTTP with stdin/stdout IPC for sr-voice workers
Gemma 2 T&C compliance: exposed HTTP ports allow mods or external code
to reach the model, complicating license enforcement. Switch to piped
stdin/stdout (JSONL protocol) so the model is only reachable through
the game server's internal queue.

- worker.rs: VoicePipe owns Child + piped stdin/stdout, VoiceProcessConfig
  replaces port-based config, workers spawn their own sr-voice child
- hardware.rs: remove VoiceInstanceManager (port/process lifecycle),
  replace with evaluate_scaling() free function + HardwareProbe::voice_config()
- sr-voice: add --stdio flag to serve command, new stdio.rs JSONL mode
- Remove ureq dependency from server crate (no longer needed)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-07 17:57:32 +01:00
jpmschweitzerandClaude Opus 4.6 1b58d8f949 feat(engine): add sr-voice LLM inference service for NPC voice pipeline
Standalone Rust crate wrapping llama-cpp-2 for GGUF model inference.
Persistent HTTP server architecture — model loaded once, requests
processed sequentially, zero CPU contention by construction.

Subcommands: serve (load model, listen), generate (single prompt),
batch (JSONL), benchmark (5-run average). Makefile targets for
build/serve/run/stop workflow.

Spike 1 validated: Gemma 2B Q4_K_M at ~16 t/s CPU, 4 cultures
tested (Krenn, Ireland, Shek'na, Aranthi), composition-engine
oath injection mechanism proven. GO for Spike 2.

Refs: D-138, #639

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-07 15:44:37 +01:00