Unable to run on Linux with Intel NPU and OpenVINO

[nilp0inter]

I have whisper.cpp with OpenVINO working on my setup and running great with CPU and GPU OpenVINO devices, but when I try to use the NPU device it fails this way:

$ whisper-cpp -m models/ggml-base.bin -f output.wav -l es --print-colors -oved NPU
whisper_init_from_file_with_params_no_state: loading model from 'models/ggml-base.bin'
whisper_init_with_params_no_state: use gpu    = 1
whisper_init_with_params_no_state: flash attn = 0
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw        = 0
whisper_init_with_params_no_state: backends   = 1
whisper_model_load: loading model
whisper_model_load: n_vocab       = 51865
whisper_model_load: n_audio_ctx   = 1500
whisper_model_load: n_audio_state = 512
whisper_model_load: n_audio_head  = 8
whisper_model_load: n_audio_layer = 6
whisper_model_load: n_text_ctx    = 448
whisper_model_load: n_text_state  = 512
whisper_model_load: n_text_head   = 8
whisper_model_load: n_text_layer  = 6
whisper_model_load: n_mels        = 80
whisper_model_load: ftype         = 1
whisper_model_load: qntvr         = 0
whisper_model_load: type          = 2 (base)
whisper_model_load: adding 1608 extra tokens
whisper_model_load: n_langs       = 99
whisper_model_load:      CPU total size =   147.37 MB
whisper_model_load: model size    =  147.37 MB
whisper_init_state: kv self size  =    6.29 MB
whisper_init_state: kv cross size =   18.87 MB
whisper_init_state: kv pad  size  =    3.15 MB
whisper_init_state: compute buffer (conv)   =   16.26 MB
whisper_init_state: compute buffer (encode) =   85.86 MB
whisper_init_state: compute buffer (cross)  =    4.65 MB
whisper_init_state: compute buffer (decode) =   96.35 MB
whisper_ctx_init_openvino_encoder_with_state: loading OpenVINO model from 'models/ggml-base-encoder-openvino.xml'
whisper_ctx_init_openvino_encoder_with_state: first run on a device may take a while ...
whisper_openvino_init: path_model = models/ggml-base-encoder-openvino.xml, device = NPU, cache_dir = models/ggml-base-encoder-openvino-cache
in openvino encoder compile routine: exception: Exception from src/inference/src/cpp/core.cpp:109:
Exception from src/inference/src/dev/plugin.cpp:53:
Exception from src/plugins/intel_npu/src/plugin/src/plugin.cpp:750:
Exception from src/plugins/intel_npu/src/compiler_adapter/src/ze_graph_ext_wrappers.cpp:361:
L0 pfnCreate2 result: ZE_RESULT_ERROR_UNKNOWN, code 0x7ffffffe - an action is required to complete the desired operation .



whisper_ctx_init_openvino_encoder_with_state: failed to init OpenVINO encoder from 'models/ggml-base-encoder-openvino.xml'

system_info: n_threads = 4 / 22 | AVX = 1 | AVX2 = 1 | AVX512 = 0 | FMA = 1 | NEON = 0 | ARM_FMA = 0 | METAL = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 0 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | CUDA = 0 | COREML = 0 | OPENVINO = 1 | CANN = 0

main: processing 'output.wav' (2135256 samples, 133.5 sec), 4 threads, 1 processors, 5 beams + best of 5, lang = es, task = transcribe, timestamps = 1 ...

I'd appreciate any insights or suggestions from the community or maintainers regarding this issue, especially in terms of fixing or debugging the problem further.

Thank you!