whisper_init_from_file_with_params_no_state: loading model from '$MODEL_DIR/ggml-large-v3-turbo-q5_0.bin'
whisper_init_with_params_no_state: use gpu    = 1
whisper_init_with_params_no_state: flash attn = 1
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw        = 0
ggml_metal_device_init: tensor API disabled for pre-M5 and pre-A19 devices
ggml_metal_library_init: using embedded metal library
ggml_metal_library_init: loaded in 0.006 sec
ggml_metal_rsets_init: creating a residency set collection (keep_alive = 180 s)
ggml_metal_device_init: GPU name:   MTL0
ggml_metal_device_init: GPU family: MTLGPUFamilyApple9  (1009)
ggml_metal_device_init: GPU family: MTLGPUFamilyCommon3 (3003)
ggml_metal_device_init: GPU family: MTLGPUFamilyMetal4  (5002)
ggml_metal_device_init: simdgroup reduction   = true
ggml_metal_device_init: simdgroup matrix mul. = true
ggml_metal_device_init: has unified memory    = true
ggml_metal_device_init: has bfloat            = true
ggml_metal_device_init: has tensor            = false
ggml_metal_device_init: use residency sets    = true
ggml_metal_device_init: use shared buffers    = true
ggml_metal_device_init: recommendedMaxWorkingSetSize  = 40200.90 MB
whisper_init_with_params_no_state: devices    = 3
whisper_init_with_params_no_state: backends   = 3
whisper_model_load: loading model
whisper_model_load: n_vocab       = 51866
whisper_model_load: n_audio_ctx   = 1500
whisper_model_load: n_audio_state = 1280
whisper_model_load: n_audio_head  = 20
whisper_model_load: n_audio_layer = 32
whisper_model_load: n_text_ctx    = 448
whisper_model_load: n_text_state  = 1280
whisper_model_load: n_text_head   = 20
whisper_model_load: n_text_layer  = 4
whisper_model_load: n_mels        = 128
whisper_model_load: ftype         = 8
whisper_model_load: qntvr         = 2
whisper_model_load: type          = 5 (large v3)
whisper_model_load: adding 1609 extra tokens
whisper_model_load: n_langs       = 100
whisper_model_load:         MTL0 total size =   573.40 MB
whisper_model_load: model size    =  573.40 MB
whisper_backend_init_gpu: device 0: MTL0 (type: 1)
whisper_backend_init_gpu: found GPU device 0: MTL0 (type: 1, cnt: 0)
whisper_backend_init_gpu: using MTL0 backend
ggml_metal_init: allocating
ggml_metal_init: found device: Apple M4 Pro
ggml_metal_init: picking default device: Apple M4 Pro
ggml_metal_init: use fusion         = true
ggml_metal_init: use concurrency    = true
ggml_metal_init: use graph optimize = true
whisper_backend_init: using BLAS backend
whisper_init_state: kv self size  =   10.49 MB
whisper_init_state: kv cross size =   31.46 MB
whisper_init_state: kv pad  size  =    7.86 MB
whisper_init_state: compute buffer (conv)   =   37.69 MB
whisper_init_state: compute buffer (encode) =   67.31 MB
whisper_init_state: compute buffer (cross)  =   63.03 MB
whisper_init_state: compute buffer (decode) =  100.04 MB

system_info: n_threads = 4 / 14 | WHISPER : COREML = 0 | OPENVINO = 0 | MTL : EMBED_LIBRARY = 1 | CPU : NEON = 1 | ARM_FMA = 1 | FP16_VA = 1 | DOTPROD = 1 | ACCELERATE = 1 | REPACK = 1 | 

main: processing '$BENCHMARK_DIR/input/vctk-p225-five-utterances-16k.wav' (404997 samples, 25.3 sec), 4 threads, 1 processors, 3 beams + best of 3, lang = auto, task = transcribe, timestamps = 0 ...

whisper_full_with_state: auto-detected language: en (p = 0.999682)
output_txt: saving output to '$BENCHMARK_DIR/runs/large-v3-turbo-q5_0-diagnostic.txt'

whisper_print_timings:     load time =   171.55 ms
whisper_print_timings:     fallbacks =   0 p /   0 h
whisper_print_timings:      mel time =     7.49 ms
whisper_print_timings:   sample time =    44.36 ms /   232 runs (     0.19 ms per run)
whisper_print_timings:   encode time =  1245.07 ms /     2 runs (   622.54 ms per run)
whisper_print_timings:   decode time =     3.28 ms /     2 runs (     1.64 ms per run)
whisper_print_timings:   batchd time =   213.89 ms /   232 runs (     0.92 ms per run)
whisper_print_timings:   prompt time =     0.00 ms /     1 runs (     0.00 ms per run)
whisper_print_timings:    total time =  1713.62 ms
ggml_metal_free: deallocating
