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

..

fixtures/gdscript

feat(ci): add make fixtures-client target for GDScript->Rust cross-encoder validation (#475)

2026-02-18 02:20:24 +01:00

golden

test(simulation): Sprint 24 tests — archetype, tell escalation, ticker, v0.1 playthrough (#593, #595)

2026-03-05 09:13:24 +01:00

integration

feat(simulation): Sprint 19 — 7 server systems

2026-02-25 12:13:03 +01:00

archetype_monologue.rs

test(simulation): Sprint 24 tests — archetype, tell escalation, ticker, v0.1 playthrough (#593, #595)

2026-03-05 09:13:24 +01:00

bridge_ipc.rs

test(simulation): Sprint 24 tests — archetype, tell escalation, ticker, v0.1 playthrough (#593, #595)

2026-03-05 09:13:24 +01:00

bridge_tcp.rs

test(simulation): Sprint 24 tests — archetype, tell escalation, ticker, v0.1 playthrough (#593, #595)

2026-03-05 09:13:24 +01:00

contamination.rs

fix(simulation): correct CONTAMINATION_DELAY_TICKS from 1800 to 300

2026-02-28 23:24:47 +01:00

content_loading.rs

feat(simulation): wire NPC pool generation and authored triangle instantiation (#176, #188)

2026-02-28 16:34:36 +01:00

content_runtime.rs

test(simulation): cross-room transition scenarios T1-T8 + test suite expansion (#506)

2026-02-19 12:04:29 +01:00

content_scaling.rs

test(simulation): Sprint 24 tests — archetype, tell escalation, ticker, v0.1 playthrough (#593, #595)

2026-03-05 09:13:24 +01:00

cross_room_transitions.rs

feat(simulation): ban HashMap via clippy disallowed_types, fix violations (#343)

2026-02-19 14:14:07 +01:00

determinism.rs

refactor(content): rename mood vocabulary to match voice guide (D-035 Sprint 14 amendment)

2026-02-20 19:00:07 +01:00

environmental_interaction.rs

fix(simulation): address PR #81 review — critical and high-priority issues

2026-03-03 23:40:59 +01:00

error_handling.rs

test(simulation): Sprint 24 tests — archetype, tell escalation, ticker, v0.1 playthrough (#593, #595)

2026-03-05 09:13:24 +01:00

fuzzy_map.rs

fix(simulation): address all round 3 PR review issues

2026-02-28 23:48:03 +01:00

game_loop.rs

feat(simulation): add NPC name masking with role labels and color index

2026-02-21 13:42:59 +01:00

gen_fixtures.rs

test(simulation): Sprint 24 tests — archetype, tell escalation, ticker, v0.1 playthrough (#593, #595)

2026-03-05 09:13:24 +01:00

gen_gauntlet_fixtures.rs

test(simulation): Sprint 24 tests — archetype, tell escalation, ticker, v0.1 playthrough (#593, #595)

2026-03-05 09:13:24 +01:00

golden_suite.rs

test(simulation): cross-room transition scenarios T1-T8 + test suite expansion (#506)

2026-02-19 12:04:29 +01:00

information_boundaries.rs

fix(simulation): address PR #81 review — critical and high-priority issues

2026-03-03 23:40:59 +01:00

ipc_bench.rs

feat(ci): protocol version handshake client + IPC benchmark (#556, #342)

2026-02-25 12:57:59 +01:00

layer3.rs

test(simulation): Sprint 24 tests — archetype, tell escalation, ticker, v0.1 playthrough (#593, #595)

2026-03-05 09:13:24 +01:00

movement.rs

feat(simulation): add entity-entity collision

2026-02-11 20:15:24 +01:00

news_ticker.rs

fix(simulation): address PR #85 review — warnings and polish items

2026-03-05 10:59:38 +01:00

perf_bench.rs

test(simulation): cross-room transition scenarios T1-T8 + test suite expansion (#506)

2026-02-19 12:04:29 +01:00

serialization.rs

test(simulation): Sprint 24 tests — archetype, tell escalation, ticker, v0.1 playthrough (#593, #595)

2026-03-05 09:13:24 +01:00

shadowcast_bench.rs

refactor(server): apply rustfmt and clippy suggestions

2026-02-16 00:40:43 +01:00

smoke.rs

test(simulation): expand test coverage per PR #2 review

2026-02-11 17:53:16 +01:00

tell_escalation.rs

fix(simulation): address PR #85 review — warnings and polish items

2026-03-05 10:59:38 +01:00

template_instantiation.rs

feat(simulation): template instantiation engine — load, spawn, lifecycle (#161)

2026-02-27 17:59:49 +01:00

template_schema.rs

feat(simulation): tier 2 template definition format — FullTemplateDef with tests (#159)

2026-02-27 17:46:37 +01:00

triangle_escalation.rs

feat(simulation): wire NPC pool generation and authored triangle instantiation (#176, #188)

2026-02-28 16:34:36 +01:00

triangle_validation.rs

feat(simulation): implement triangle validation and cross-template generation (#108, #109)

2026-02-27 11:11:31 +01:00

v01_integration_playthrough.rs

test(simulation): Sprint 24 tests — archetype, tell escalation, ticker, v0.1 playthrough (#593, #595)

2026-03-05 09:13:24 +01:00

voice_pipeline.rs

feat(voice): complete Spike 2 voice pipeline with quality-tested prompt engine

2026-03-07 19:20:06 +01:00

voice_quality_batch.rs

feat(voice): complete Spike 2 voice pipeline with quality-tested prompt engine

2026-03-07 19:20:06 +01:00