Skip to content

Commit 0048fe1

Browse files
committed
OKFS+papers: ADR-KAI4 THE LATENT VOICE (P0 shipped, Audex teardown, text-bypass roadmap)
1 parent 99ab42b commit 0048fe1

4 files changed

Lines changed: 139 additions & 0 deletions

File tree

memory-okf/LUT.md

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -202,3 +202,4 @@ Lookup before you build. `python tools/okf_mem.py lookup --root <root> <kw>`
202202
| c88997050745eb52 | agent | live-play round2,memory pollution,tool_output captured,admission hygiene,fence mutations,unclosed fence,prose-then-fence,autoroute run_python,fuzzy tool resolve,spine recall qonly,tool discipline residual,hallucinated weather | B4 captured tool_output/greetings as memories (fix: admission hygiene engine 6e02043 + registry cleaned 31->7 + spine recall QONLY). Fence-mutation census all parsed (kairos d95faad): case/unclosed/mashup/name-split/prose-then-fence + python-block autoroute via sandboxed run_python + fuzzy name resolve. Replays: greet 1.3-2.9s, get_time correct 16.9s, name 4.9s. RESIDUAL G-TOOL-DISCIPLINE: reason-SFT model freestyles format + sometimes answers without calling (hallucinated weather) - needs tool-use SFT or temp clamp. | GREEN | sum/c88997050745eb52.md |
203203
| c2b4f1e275bd420a | agent | live-play round3,Tool-Code hyphen,stream malformed recovery,persona internal dials,neutral recite fix,registry round3 clean,butterfly memories kept | Tool-Code hyphen parsed; stream re-prompts on malformed fences (never flushes raw); personality rendered as internal dials (no more literal 'Neutral.'); junk pruned, genuine conversational memories kept. Gates hold. | GREEN | sum/c2b4f1e275bd420a.md |
204204
| c8b97c157c1d9c9e | agent | live-play 4,toolset swap aborted,system prompt stability law,spine_toolset off,coding tools index,token 0 divergence,one stable system prompt | SP_SPINE_TOOLSET per-turn swap rewrote the system prompt -> persist cache diverged at token 0 -> whole-conversation re-prefill on any 'build/code' mention. Profile: spine_toolset=false; coding tools now in the all_tools load-on-demand index (deduped). Adaptive things go in the message stream or tool index, never the system prefix. | GREEN | sum/c8b97c157c1d9c9e.md |
205+
| fdebbd495a530e25 | agent | KAI-4,latent voice,GNA ear,inject_frames,audio placeholder 258881,Audex teardown,discrete token reject,voice head,GAN vocoder,V_sub vocab,G-VOICE-0,voice channel GUI,voxtral reuse,E=3840,WSL no GNA MMIO | Sovereign voice: KAI-3 GNA ear (silicon-gated) wired to mic + /v1/voice + console -> inject_frames -> 12B (G-VOICE-0 parity 8/8, live e2e proven). Audex: steal unified-input + 2-stage output split, reject discrete-token output. P0=plumbing GREEN; P1 vocab scale-up makes free speech legible; P3 voice head = text bypass. E=3840 not 1024; GNA is frozen INT16 conv-only leverage. | GREEN | sum/fdebbd495a530e25.md |
Lines changed: 23 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,23 @@
1+
---
2+
type: memory
3+
title: ADR-KAI4 THE LATENT VOICE: GNA ear P0 shipped (G-VOICE-0 8/8), Audex teardown, text-bypass roadmap
4+
description: Sovereign voice: KAI-3 GNA ear (silicon-gated) wired to mic + /v1/voice + console -> inject_frames -> 12B (G-VOICE-0 parity 8/8, live e2e proven). Audex: steal unified-input + 2-stage output split, reject discrete-token output. P0=plumbing GREEN; P1 vocab scale-up makes free speech legible; P3 voice head = text bypass. E=3840 not 1024; GNA is frozen INT16 conv-only leverage.
5+
timestamp: 2026-07-10T15:26:18Z
6+
resource: kairos e180cea; harness 1dedacb; engine 98f8313
7+
sp_status: GREEN
8+
sp_gate: G-VOICE-0 parity 8/8 + live wav->ear->inject e2e
9+
sp_commit: kairos e180cea; harness 1dedacb; engine 98f8313
10+
sp_repro: serve.bat agent; python harness_tests/g_voice0_parity.py; console voice button
11+
mem_kind: agent
12+
mem_addr: fdebbd495a530e25
13+
mem_verified: unverified
14+
mem_lifecycle: active
15+
tags: [KAI-4, latent voice, GNA ear, inject_frames, audio placeholder 258881, Audex teardown, discrete token reject, voice head, GAN vocoder, V_sub vocab, G-VOICE-0, voice channel GUI, voxtral reuse, E=3840, WSL no GNA MMIO, agent, tier-2]
16+
mem_tier: full
17+
---
18+
19+
ADR-KAI4 THE LATENT VOICE (2026-07-11, operator voice campaign). GOAL: sovereign realtime bidirectional voice - straight into the 12B's residual audio channel, text bypassed where possible, offloaded to the deprecated Intel GNA 2.0 (frozen target = leverage).
20+
INVENTORY (do not re-derive): the INPUT half is BUILT + GATED ON SILICON. KAI-3 ear = log-mel(64, 40ms/640 hop, 16k) -> Conv1d*3(256) CTC encoder -> POT i16 GNA-legal (pad=0, head padded to 36ch/÷4) -> CTC-greedy -> softmax(tau=0.2)*W_sub(√H-scaled embed rows) -> [k×E=3840] on-manifold frames -> daemon inject_frames + audio placeholder 258881 via gemma4_kv_inject_seq. Gated: 7/8 pivot, i16==FP32 0.877, GNA_HW bit-parity (G-KAIROS-3-GNA-HW). ckpt C:\Users\Knack\audio_ctc.pt. Reuse: OUR voxtral-mini-realtime-rs (hound/rubato/rustfft/ring_buffer + L0 iGPU + flow-matching TTS). ★WSL2 has NO GNA MMIO - live pipeline is Windows-native.
21+
AUDEX-2B TEARDOWN (arXiv 2607.05196, actually a 30B-A3B MoE + a 2B; NVIDIA one-way NONCOMMERCIAL license so we don't ship their weights): STEAL = unified continuous-input-projection (validates our inject_frames seam, we gated it first) + the 2-STAGE OUTPUT SPLIT (even NVIDIA doesn't emit waveforms from the LLM: compact acoustic stream -> small causal/streaming decoder = exactly our Voice-Head->GAN) + <sound> placeholder/template discipline. REJECT = discrete audio tokens through the text softmax (vocab extension + RVQ + token-rate decode on the 2060 = the tokenization tax we exist to avoid; our output leaves as a CONTINUOUS conditioning vector, no backbone surgery) + their 157B-token encoder economics + GPU-resident audio I/O.
22+
GEMINI CORRECTIONS: manifold is E=3840 NOT 1024; ear is already trained+on-silicon (work is STREAMING it not building it); we serve the 12B not a 9B; GNA is inference-only INT16 conv/affine no-transpose-conv no-padding ch÷4 (vocoder must pass a POT gate like the ear, don't assume); iGPU/L0 is a 4th compute tier Gemini missed. Compute map: GNA=always-on ear+wake, iGPU=heavier audio nets, CPU=vocoder v0, 2060=100% Gemma.
23+
P0 SHIPPED (kairos e180cea): harness/voice (dsp.py VERBATIM from gen_audio_frames.py trainer; ear.py OV GNA->CPU ladder; service.py); gateway /v1/voice + /v1/voice/status; console voice channel (mic getUserMedia 16k -> client VAD segmentation -> PCM16 b64 -> /v1/voice -> GNA ear -> inject_frames -> streamed reply -> v0 speechSynthesis TTS; push-to-talk/auto/interrupt-on-speak-while-speaking); tools/voice_export_wsub.py (torch-free W_sub from safetensors bf16<<16). G-VOICE-0 PARITY 8/8 (live ear == gated KAI2 packets, cos 1.0000, device CPU - GNA_HW needs the driver live in this process). LIVE E2E proven: 6.72s wav -> ear 4 frames -> daemon inject (S1 prompt injected). HONEST P0 boundary: reply empty on arbitrary speech - the ear's legible vocab is only V_sub=32 (KAI-3 action/no-op corpus); P0 gates PLUMBING, P1 = vocab scale-up 32->512 (multi-voice TTS render via our crate) + retrain + GNA re-gate. Ladder: P1 ear grows + wake head; P2 our CPU-GAN vocoder replaces browser TTS; P3 VOICE HEAD (hidden->conditioning, trained on self-generated pairs) = text-bypass leap; P4 GNA vocoder port. Gates G-VOICE-0..LIVE in the ADR.

memory-okf/sum/fdebbd495a530e25.md

Lines changed: 24 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,24 @@
1+
---
2+
type: memory
3+
title: ADR-KAI4 THE LATENT VOICE: GNA ear P0 shipped (G-VOICE-0 8/8), Audex teardown, text-bypass roadmap
4+
description: Sovereign voice: KAI-3 GNA ear (silicon-gated) wired to mic + /v1/voice + console -> inject_frames -> 12B (G-VOICE-0 parity 8/8, live e2e proven). Audex: steal unified-input + 2-stage output split, reject discrete-token output. P0=plumbing GREEN; P1 vocab scale-up makes free speech legible; P3 voice head = text bypass. E=3840 not 1024; GNA is frozen INT16 conv-only leverage.
5+
timestamp: 2026-07-10T15:26:18Z
6+
resource: kairos e180cea; harness 1dedacb; engine 98f8313
7+
sp_status: GREEN
8+
sp_gate: G-VOICE-0 parity 8/8 + live wav->ear->inject e2e
9+
sp_commit: kairos e180cea; harness 1dedacb; engine 98f8313
10+
sp_repro: serve.bat agent; python harness_tests/g_voice0_parity.py; console voice button
11+
mem_kind: agent
12+
mem_addr: fdebbd495a530e25
13+
mem_verified: unverified
14+
mem_lifecycle: active
15+
tags: [KAI-4, latent voice, GNA ear, inject_frames, audio placeholder 258881, Audex teardown, discrete token reject, voice head, GAN vocoder, V_sub vocab, G-VOICE-0, voice channel GUI, voxtral reuse, E=3840, WSL no GNA MMIO, agent, tier-1]
16+
mem_tier: summary
17+
mem_full: fdebbd495a530e25
18+
---
19+
20+
# ADR-KAI4 THE LATENT VOICE: GNA ear P0 shipped (G-VOICE-0 8/8), Audex teardown, text-bypass roadmap
21+
22+
Sovereign voice: KAI-3 GNA ear (silicon-gated) wired to mic + /v1/voice + console -> inject_frames -> 12B (G-VOICE-0 parity 8/8, live e2e proven). Audex: steal unified-input + 2-stage output split, reject discrete-token output. P0=plumbing GREEN; P1 vocab scale-up makes free speech legible; P3 voice head = text bypass. E=3840 not 1024; GNA is frozen INT16 conv-only leverage.
23+
24+
Full context: [full/fdebbd495a530e25.md](../full/fdebbd495a530e25.md)

0 commit comments

Comments
 (0)