3.1 Speech-to-text: whisper.cpp - PASS, 28x real-time

• Build: whisper.cpp master @ clone 2026-07-04, -DGGML_VULKAN=ON, checkout at ~/sparkbench/whisper.cpp (gitignored). Built clean, first try.
• Vulkan acceleration CONFIRMED: model loaded to Vulkan0 (Radeon 8060S RADV GFX1151), 1.62GB on-device for large-v3-turbo.
• Test audio: samples/jfk.wav (11s) + a 330s file made by concatenating it 30x (stdlib wave; no ffmpeg on host - see DEFERRED-ROOT note below).

• Finding: quality is free. small and large-v3-turbo run at the SAME speed on this stack (both ~28x; the bottleneck is not the encoder GPU work at these sizes). Always use large-v3-turbo.
• Consulting question ("real-time meeting transcription?"): YES, emphatically. A 1-hour meeting transcribes in ~2 minutes with the best turbo model; live streaming use leaves >25x headroom.
• Caveats: single-speaker clean audio sample; repeated content may flatter the decoder slightly. Real meeting audio (crosstalk, accents) is a quality question for a later eval, not a throughput one. ffmpeg absent (apt) - listed in DEFERRED-ROOT.md alongside ROCm; stdlib wave sufficed here.

3.2 Image generation: stable-diffusion.cpp - PASS (with one required flag)

• Build: stable-diffusion.cpp master @ clone 2026-07-04, -DSD_VULKAN=ON, at ~/sparkbench/stable-diffusion.cpp (gitignored). Built clean. NOTE: the CLI binary is now sd-cli, not sd (briefing's command is stale); FLUX split files load via --diffusion-model, not -m.
• Model set (tool-local models/, provenance = size+sha256 verified at fetch): flux1-schnell-Q4_0.gguf (6.77GB, city96, oid-matched); ae.safetensors (335MB - SUBSTITUTION: black-forest-labs repo is GATED (401), ungated second-state mirror used, byte-size identical, oid-matched); clip_l (246MB) + t5xxl_fp16 (9.79GB) from comfyanonymous, oid-matched.
• VRAM: 15.94GB total (t5xxl 9.3GB dominates - fp8 variant would halve this if VRAM ever matters).

• FINDING - deployment gotcha: at 1024x1024 the FLUX VAE requests an 8.5GB contiguous compute buffer; RADV's single-allocation limit (~4.3GB) rejects it AFTER 40s of successful sampling. --vae-tiling is mandatory at 1024 on this stack (7.2s tiled decode, no quality complaint visible). Failure mode is late and expensive - bake the flag into any wrapper.
• Quality: both images coherent, well-lit, artefact-free to my eye; sent to Alastair 2026-07-04 for the subjective call.
• Consulting question ("local images instead of Midjourney/ DALL-E?"): qualified YES - 9.5s at 512, 48s at 1024, on-prem, no per-image billing; the qualification is FLUX-schnell's style range vs commercial tools, a quality question not a speed one.

3.2 addendum: 50-image style gallery + resolution ceiling (2026-07-04)

• Gallery batch (Alastair request): 50 business-blog prompts across 10 style families (photography, flat/isometric illustration, watercolor, oil, cartoon/comic, 3D render, sketch/line art, pixel art, business-concept), 1024x1024, seeds 1001-1050, prompts tracked in results/flux-gallery-prompts.txt. 50/50 succeeded, ~36s sampling each, ~55 min wall total, images in results/flux-gallery/ (gitignored, 67MB) + zip delivered to Alastair. Sampling time was style-invariant (35.9-36.1s across all families) - cost depends on pixels, never on content or style.
• Resolution ceiling (Alastair request - 1080p/4K):

• FINDING: the binding constraint is the RADV single-allocation cap (~4.3GB) on the transformer's attention buffer, which grows ~quadratically with pixel count - NOT total VRAM (only 16GB of 110GB was in use). Practical ceiling ~1.3-1.5MP. Production recipe for 1080p/4K deliverables: generate at 1536x864 and upscale (1.25x to 1080p; a dedicated upscaler for 4K - candidate future probe, e.g. realesrgan-ncnn-vulkan). Failure mode caveat from the first attempt: errors can arrive AFTER a full sampling phase, and a filtered log can hide them - always check the output file exists.

3.3 Text-to-speech: Piper - PASS, 26.6x real-time on CPU

• Install: pre-built binary release 2023.11.14-2 (pip unavailable on host - python3-pip in DEFERRED-ROOT.md; the briefing's binary fallback used instead). Location: ~/sparkbench/piper/ (gitignored); tarball sha256 recorded in docs/TOOLCHAIN.md.
• Voice: en_GB-alba-medium (63MB onnx, sha256 in TOOLCHAIN.md) - British English per the briefing's Irish-market note.
• Test: 497-character paragraph -> 30.1s of audio in 1.13s = 26.6x real-time, CPU-only (no GPU involvement at all - this leaves the GPU free for concurrent LLM work).
• Quality: APPROVED - Alastair auditioned the sample 2026-07-04: "sounded great", and he specifically likes the alba voice. en_GB-alba-medium is the house default voice going forward.
• Consulting question ("local speech generation for accessibility/training?"): YES, unqualified - a 10-minute narration renders in ~23 seconds on CPU while the GPU serves LLM load, and the quality passed the founder's ear test.