Kazeia-engine/dist/jni
Richard Loyer 3775bbc37c chantier TTS CosyVoice3 + verdict R&D NPU
dist/cosyvoice/ : engine streaming chunk-causal (RTF 0.88, cache prompt block-causal,
  HiFT F16) + clonage vocal ECAPA + JNI + libs + exemple voix (damien.cvps)
dist/distill/ : pipeline distillation Reflow 2-NFE (train, export ONNX, eval)
dist/jni/ : CosyVoiceTtsEngine.kt + kazeia_cosyvoice_jni + maj engine/tts/speaker_encoder
dist/lib-cpu/ : libs ggml/llama CPU-only (contournement SELinux FastRPC untrusted_app)
dist/outetts/ : runner OuteTTS (écarté au profit de CosyVoice3)

R&D NPU (détail en mémoire) : QNN HTP exécute MAL le flow DiT de façon irréductible
hors-root (CPU-backend fidèle 0.74 vs HTP 0.14, même DLC ; chirurgie ONNX canonicalisée,
opt-level sans effet). Hexagon-MLIR open exécute GPT-2 correctement sur V79 -> QNN
backend coupable, pas le silicium. Ship = CPU RTF 0.88 ; bug-report Qualcomm = vrai levier.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-16 21:48:30 +02:00
..
CosyVoiceTtsEngine.kt chantier TTS CosyVoice3 + verdict R&D NPU 2026-06-16 21:48:30 +02:00
EngineLlmEngine.kt chantier TTS CosyVoice3 + verdict R&D NPU 2026-06-16 21:48:30 +02:00
SttEngine.kt chantier B STT #2 : S2 + S3 livrés (stt_engine ORT QNN + JNI + façade Kotlin + doc) 2026-05-31 22:15:58 +02:00
TtsEngine.kt TU sampler tuning : top_p 1.0->0.95 + rep_penalty 1.05->1.10 nouveaux défauts 2026-05-31 11:18:25 +02:00
build_prefill.cpp chantier B TTS #4: P3.1+P3.2 = x_vector + prefill_embeds depuis input_ids (bit-exact) 2026-05-28 15:05:23 +02:00
cp_inference.cpp chantier B engine #3 : TTS lib audible + E2E in-app validé 2026-06-03 22:37:47 +02:00
cp_inference.h chantier F (CP HMX) : path sched livré, gain absent (overhead per-call) 2026-05-29 16:03:41 +02:00
cp_validate.cpp chantier B TTS #5: P3.3 = vrai CP runner intégré (bit-exact) 2026-05-28 15:11:07 +02:00
decoder_subprocess.cpp Decoder chraac en sous-process : -20% decoder, -9% pipeline, talker intact 2026-05-30 18:14:09 +02:00
dual_ctx.cpp dist: prefill HTP = CASSÉ (sortie charabia), prefill obligatoirement CPU 2026-05-27 09:42:59 +02:00
dual_ctx_mt.cpp dist: JNI option C câblé (prefill HTP / decode CPU), validé multi-tour 2026-05-27 10:24:55 +02:00
kazeia_cosyvoice_jni.cpp chantier TTS CosyVoice3 + verdict R&D NPU 2026-06-16 21:48:30 +02:00
kazeia_engine_jni.cpp chantier TTS CosyVoice3 + verdict R&D NPU 2026-06-16 21:48:30 +02:00
kazeia_mel.cpp chantier B STT #6 : 2 bugs critiques fixés grâce à l'A/B in-app du dev 2026-06-01 12:19:33 +02:00
kazeia_mel.h chantier B STT #1 : S1 unification engine (mel partagé + skeleton stt + VAD C++) 2026-05-31 19:19:28 +02:00
kazeia_stt_jni.cpp chantier B STT #2 : S2 + S3 livrés (stt_engine ORT QNN + JNI + façade Kotlin + doc) 2026-05-31 22:15:58 +02:00
kazeia_text_tokenizer.cpp chantier B TTS #10: tokenizer Qwen3 BPE via llama_tokenize, texte arbitraire 2026-05-28 21:21:53 +02:00
kazeia_text_tokenizer.h chantier B TTS #10: tokenizer Qwen3 BPE via llama_tokenize, texte arbitraire 2026-05-28 21:21:53 +02:00
kazeia_tts_jni.cpp chantier TTS CosyVoice3 + verdict R&D NPU 2026-06-16 21:48:30 +02:00
sampler.cpp chantier B TTS #7: sampler HF-style top_k+top_p+rep_penalty (générique) 2026-05-28 16:55:05 +02:00
sampler.h chantier B TTS #7: sampler HF-style top_k+top_p+rep_penalty (générique) 2026-05-28 16:55:05 +02:00
speaker_encoder.cpp chantier TTS CosyVoice3 + verdict R&D NPU 2026-06-16 21:48:30 +02:00
speaker_encoder.h chantier B TTS #8 : speaker encoder embedded + clonage in-process 2026-05-31 15:54:29 +02:00
stt_cli.cpp chantier B STT #6 : 2 bugs critiques fixés grâce à l'A/B in-app du dev 2026-06-01 12:19:33 +02:00
stt_engine.cpp chantier B STT #6 : 2 bugs critiques fixés grâce à l'A/B in-app du dev 2026-06-01 12:19:33 +02:00
stt_engine.h chantier B STT #1 : S1 unification engine (mel partagé + skeleton stt + VAD C++) 2026-05-31 19:19:28 +02:00
stt_test.cpp chantier B STT #1 : S1 unification engine (mel partagé + skeleton stt + VAD C++) 2026-05-31 19:19:28 +02:00
test_cp_iso.cpp chantier TTS CosyVoice3 + verdict R&D NPU 2026-06-16 21:48:30 +02:00
test_engine.c Valide device: generate() bout-en-bout OK (test_native), bridge pret integration 2026-05-24 22:05:44 +02:00
test_engine.cpp Valide device: generate() bout-en-bout OK (test_native), bridge pret integration 2026-05-24 22:05:44 +02:00
test_engine_2calls.cpp chantier B TTS #11: libkazeia_tts.so + Kotlin wrapper, JNI validé bout-en-bout 2026-05-28 21:54:21 +02:00
test_jni_native.cpp dist: multi-tour propre (generateRaw + ChatSession) + JNI validé end-to-end natif 2026-05-27 11:18:20 +02:00
test_jni_tts.cpp chantier B engine #3 : TTS lib audible + E2E in-app validé 2026-06-03 22:37:47 +02:00
test_native.cpp v4 fix #277: ngl0 CPU decode 14tok/s, 4s/tour, usable 2026-05-25 10:18:31 +02:00
test_talker.cpp chantier B TTS #1: API JNI embeds-only + M-RoPE qwen3 dense 2026-05-28 13:37:17 +02:00
test_talker_replay.cpp chantier B TTS #2: fix M-RoPE positions en embeds-mode + validation bit-exact 2026-05-28 13:54:03 +02:00
test_tokenizer.cpp chantier B TTS #10: tokenizer Qwen3 BPE via llama_tokenize, texte arbitraire 2026-05-28 21:21:53 +02:00
tts_engine.cpp chantier B engine #3 : TTS lib audible + E2E in-app validé 2026-06-03 22:37:47 +02:00
tts_engine.h chantier B TTS #8 : speaker encoder embedded + clonage in-process 2026-05-31 15:54:29 +02:00
tts_orchestrate.cpp chantier B TTS #6: P3.4+P3.5 = pipeline complet 100% tablette bout-en-bout 2026-05-28 15:25:00 +02:00
tts_pipeline.cpp chantier B STT #1 : S1 unification engine (mel partagé + skeleton stt + VAD C++) 2026-05-31 19:19:28 +02:00