Ettore Di Giacinto 487a2a3dc8 backend(audio-cpp): load models and cache sessions per task ...

Loads one ILoadedVoiceModel and creates an IVoiceTaskSession lazily per
(task, mode), so the same model serves both the unary and streaming RPCs.
LoadModel derives the family from GGUF metadata or an explicit option and
fails with INVALID_ARGUMENT otherwise, so a failed load is a gRPC error the
backend probe can see.

audiocpp_backend::Task mirrors engine::runtime::VoiceTaskKind positionally,
and drift there is silent: every unit still compiles and every test still
passes while the backend runs a different task. Two mechanisms pin it. The
static_asserts in loaded_model.cpp catch an insertion or a reorder, and
-Werror=switch on that one file turns an appended upstream enumerator into a
build failure rather than a warning in a 600 file log.

Assisted-by: Claude:claude-opus-5 [Claude Code]
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

2026-07-29 19:03:32 +00:00

..

.gitignore

backend(audio-cpp): add the native build scaffold

2026-07-29 19:03:32 +00:00

audio_units_test.cpp

backend(audio-cpp): map NaN PCM samples to silence explicitly

2026-07-29 19:03:32 +00:00

audio_units.cpp

backend(audio-cpp): map NaN PCM samples to silence explicitly

2026-07-29 19:03:32 +00:00

audio_units.h

backend(audio-cpp): map NaN PCM samples to silence explicitly

2026-07-29 19:03:32 +00:00

capability_routing_test.cpp

backend(audio-cpp): use upstream's 'spk' task name and pin the preference order

2026-07-29 19:03:32 +00:00

capability_routing.cpp

backend(audio-cpp): use upstream's 'spk' task name and pin the preference order

2026-07-29 19:03:32 +00:00

capability_routing.h

backend(audio-cpp): use upstream's 'spk' task name and pin the preference order

2026-07-29 19:03:32 +00:00

CMakeLists.txt

backend(audio-cpp): load models and cache sessions per task

2026-07-29 19:03:32 +00:00

family_gate_test.cpp

backend(audio-cpp): gate model loading on the audio.cpp family

2026-07-29 19:03:32 +00:00

family_gate.cpp

backend(audio-cpp): gate model loading on the audio.cpp family

2026-07-29 19:03:32 +00:00

family_gate.h

backend(audio-cpp): gate model loading on the audio.cpp family

2026-07-29 19:03:32 +00:00

grpc-server.cpp

backend(audio-cpp): load models and cache sessions per task

2026-07-29 19:03:32 +00:00

inference_lane_test.cpp

backend(audio-cpp): make the B10 test able to fail, and document LaneEntry

2026-07-29 19:03:32 +00:00

inference_lane.cpp

backend(audio-cpp): serialize inference through an InferenceLane

2026-07-29 19:03:32 +00:00

inference_lane.h

backend(audio-cpp): make the B10 test able to fail, and document LaneEntry

2026-07-29 19:03:32 +00:00

loaded_model.cpp

backend(audio-cpp): load models and cache sessions per task

2026-07-29 19:03:32 +00:00

loaded_model.h

backend(audio-cpp): load models and cache sessions per task

2026-07-29 19:03:32 +00:00

Makefile

backend(audio-cpp): keep the build-tree rpath at $ORIGIN

2026-07-29 19:03:32 +00:00

model_options_test.cpp

backend(audio-cpp): reject out-of-range numeric model options

2026-07-29 19:03:32 +00:00

model_options.cpp

backend(audio-cpp): reject out-of-range numeric model options

2026-07-29 19:03:32 +00:00

model_options.h

backend(audio-cpp): parse namespaced model options

2026-07-29 19:03:32 +00:00

transcript_assembly_test.cpp

backend(audio-cpp): keep a nested speaker turn's own label

2026-07-29 19:03:32 +00:00

transcript_assembly.cpp

backend(audio-cpp): keep a nested speaker turn's own label

2026-07-29 19:03:32 +00:00

transcript_assembly.h

backend(audio-cpp): keep a nested speaker turn's own label

2026-07-29 19:03:32 +00:00