demo_vllm.py执行结果为空

[AllenMac]

FunASR Version

1.3.5

Python Version

3.10.12

Device

cuda (GPU)

Bug Description

使用python demo_vllm.py --input zh.mp3 --language 中文 --tensor-parallel-size 1 --model-dir /app/models/Fun-ASR-Nano-2512 --dtype fp16 --gpu-memory-utilization 0.5执行结果为空

Steps to Reproduce

python demo_vllm.py --input zh.mp3 --language 中文 --tensor-parallel-size 1 --model-dir /app/models/Fun-ASR-Nano-2512 --dtype fp16 --gpu-memory-utilization 0.5


gpu:
Driver Version: 570.158.01  
Tesla V100-SXM2-32GB

python虚拟环境:
vllm==0.12.0
funasr==1.3.5


Package                           Version
--------------------------------- -------------
aiohappyeyeballs                  2.6.2
aiohttp                           3.13.5
aiosignal                         1.4.0
aliyun-python-sdk-core            2.16.0
aliyun-python-sdk-kms             2.16.5
annotated-doc                     0.0.4
annotated-types                   0.7.0
anthropic                         0.71.0
antlr4-python3-runtime            4.9.3
anyio                             4.13.0
apache-tvm-ffi                    0.1.11
astor                             0.8.1
async-timeout                     5.0.1
attrs                             26.1.0
audioread                         3.1.0
backports.strenum                 1.3.1
blake3                            1.0.8
cachetools                        7.1.4
cbor2                             6.1.1
certifi                           2026.5.20
cffi                              2.0.0
charset-normalizer                3.4.7
click                             8.4.1
cloudpickle                       3.1.2
compressed-tensors                0.12.2
crcmod                            1.7
cryptography                      48.0.0
cuda-bindings                     13.3.1
cuda-core                         1.0.1
cuda-pathfinder                   1.5.5
cuda-python                       13.3.1
cupy-cuda12x                      14.1.1
decorator                         5.3.1
depyf                             0.20.0
detect-installer                  0.1.0
dill                              0.4.1
diskcache                         5.6.3
distro                            1.9.0
dnspython                         2.8.0
docstring_parser                  0.18.0
editdistance                      0.8.1
einops                            0.8.2
email-validator                   2.3.0
exceptiongroup                    1.3.1
fastapi                           0.136.3
fastapi-cli                       0.0.24
fastapi-cloud-cli                 0.19.0
fastar                            0.11.0
filelock                          3.29.0
flashinfer-python                 0.5.3
frozenlist                        1.8.0
fsspec                            2026.4.0
funasr                            1.3.5
gguf                              0.19.0
h11                               0.16.0
hf-xet                            1.5.0
httpcore                          1.0.9
httptools                         0.8.0
httpx                             0.28.1
huggingface_hub                   0.36.2
hydra-core                        1.3.2
idna                              3.17
interegular                       0.3.3
jaconv                            0.5.0
jamo                              0.4.1
jieba                             0.42.1
Jinja2                            3.1.6
jiter                             0.15.0
jmespath                          0.10.0
joblib                            1.5.3
jsonschema                        4.26.0
jsonschema-specifications         2025.9.1
kaldiio                           2.18.1
lark                              1.2.2
lazy-loader                       0.5
librosa                           0.11.0
llguidance                        1.3.0
llvmlite                          0.44.0
lm-format-enforcer                0.11.3
loguru                            0.7.3
markdown-it-py                    4.2.0
MarkupSafe                        3.0.3
mdurl                             0.1.2
mistral_common                    1.11.2
model-hosting-container-standards 0.1.15
modelscope                        1.37.1
mpmath                            1.3.0
msgpack                           1.1.2
msgspec                           0.21.1
multidict                         6.7.1
networkx                          3.4.2
ninja                             1.13.0
numba                             0.61.2
numpy                             2.2.6
nvidia-cublas-cu12                12.8.4.1
nvidia-cuda-cupti-cu12            12.8.90
nvidia-cuda-nvrtc-cu12            12.8.93
nvidia-cuda-runtime-cu12          12.8.90
nvidia-cudnn-cu12                 9.10.2.21
nvidia-cudnn-frontend             1.24.0
nvidia-cufft-cu12                 11.3.3.83
nvidia-cufile-cu12                1.13.1.3
nvidia-curand-cu12                10.3.9.90
nvidia-cusolver-cu12              11.7.3.90
nvidia-cusparse-cu12              12.5.8.93
nvidia-cusparselt-cu12            0.7.1
nvidia-cutlass-dsl                4.5.2
nvidia-cutlass-dsl-libs-base      4.5.2
nvidia-ml-py                      13.595.45
nvidia-nccl-cu12                  2.27.5
nvidia-nvjitlink-cu12             12.8.93
nvidia-nvshmem-cu12               3.3.20
nvidia-nvtx-cu12                  12.8.90
omegaconf                         2.3.0
openai                            2.38.0
openai-harmony                    0.0.8
opencv-python-headless            4.13.0.92
oss2                              2.19.1
outlines_core                     0.2.11
packaging                         26.2
partial-json-parser               0.2.1.1.post7
pillow                            12.2.0
pip                               22.0.2
platformdirs                      4.10.0
pooch                             1.9.0
prometheus_client                 0.25.0
prometheus-fastapi-instrumentator 8.0.0
propcache                         0.5.2
protobuf                          7.35.0
psutil                            7.2.2
py-cpuinfo                        9.0.0
pybase64                          1.4.3
pycountry                         26.2.16
pycparser                         3.0
pycryptodome                      3.23.0
pydantic                          2.13.4
pydantic_core                     2.46.4
pydantic-extra-types              2.11.1
pydantic-settings                 2.14.1
Pygments                          2.20.0
pynndescent                       0.6.0
python-dotenv                     1.2.2
python-json-logger                4.1.0
python-multipart                  0.0.30
pytorch-wpe                       0.0.1
PyYAML                            6.0.3
pyzmq                             27.1.0
ray                               2.55.1
referencing                       0.37.0
regex                             2026.5.9
requests                          2.34.2
rich                              15.0.0
rich-toolkit                      0.19.10
rignore                           0.7.6
rpds-py                           0.30.0
safetensors                       0.7.0
scikit-learn                      1.7.2
scipy                             1.15.3
sentencepiece                     0.2.1
sentry-sdk                        2.61.1
setproctitle                      1.3.7
setuptools                        59.6.0
shellingham                       1.5.4
six                               1.17.0
sniffio                           1.3.1
soundfile                         0.13.1
soxr                              1.1.0
starlette                         1.2.1
supervisor                        4.3.0
sympy                             1.14.0
tabulate                          0.10.0
tensorboardX                      2.6.5
threadpoolctl                     3.6.0
tiktoken                          0.13.0
tokenizers                        0.22.2
tomli                             2.4.1
torch                             2.9.0
torch-complex                     0.4.4
torchaudio                        2.9.0
torchvision                       0.24.0
tqdm                              4.67.3
transformers                      4.57.6
triton                            3.5.0
typer                             0.26.4
typing_extensions                 4.15.0
typing-inspection                 0.4.2
umap-learn                        0.5.12
urllib3                           2.7.0
uvicorn                           0.48.0
uvloop                            0.22.1
vllm                              0.12.0
watchfiles                        1.2.0
websockets                        16.0
xgrammar                          0.1.27
yarl                              1.24.2

Error Message / Traceback

[LauraGPT]

你好！请尝试以下排查：

1. 升级 FunASR 到最新版：

pip install --upgrade funasr

当前最新版 v1.3.9 修复了多个 vLLM 兼容性问题。

2. 检查 vLLM 版本：Fun-ASR-Nano 推荐 vLLM >= 0.8.0。确认你的 vLLM 版本：

pip show vllm

3. 检查模型路径：确保 /app/models/Fun-ASR-Nano-2512 包含完整的模型文件。也可以尝试使用 HuggingFace 自动下载：

python demo_vllm.py --input zh.mp3 --language 中文 --model-dir FunAudioLLM/Fun-ASR-Nano-2512

4. 检查音频格式：确认 zh.mp3 是有效的音频文件，可以用 ffprobe zh.mp3 验证。

如果问题仍然存在，请贴出完整的终端输出（包括模型加载过程）。

[AllenMac]

1.已经升级为1.3.9,问题未解决
2.vllm版本如前面描述，是0.12.0

3.模型是通过modelscope命令下载下来的，为了验证，已经重新执行python demo_vllm.py --input zh.mp3 --language 中文 --model-dir FunAudioLLM/Fun-ASR-Nano-2512 命令重新下载，执行这个命令会由于V100不支持bf16报错，改为python demo_vllm.py --input zh.mp3 --language 中文 --model-dir FunAudioLLM/Fun-ASR-Nano
-2512 --dtype fp16 --gpu-memory-utilization 0.5 ，依旧没有结果输出，error为Cannot use FA version 2 is not supported due to FA2 is only supported on devices with compute capability >= 8

，详细日志如下：

点击查看完整日志（共103行）

============================================================
Fun-ASR-Nano vLLM Inference
============================================================
  Model: FunAudioLLM/Fun-ASR-Nano-2512
  Tensor Parallel: 1 GPU(s)
  Dtype: fp16
  Language: 中文
  Hotwords: (none)

Downloading Model from https://www.modelscope.cn to directory: /root/.cache/modelscope/hub/models/FunAudioLLM/Fun-ASR-Nano-2512
INFO 06-03 19:28:58 [utils.py:253] non-default args: {'enable_prompt_embeds': True, 'trust_remote_code': True, 'dtype': 'float16', 'seed': None, 'max_model_len': 2048, 'gpu_memory_utilization': 0.5, 'disable_log_stats': True, 'model': '/root/.cache/modelscope/hub/models/FunAudioLLM/Fun-ASR-Nano-2512/Qwen3-0.6B-vllm'}
The argument `trust_remote_code` is to be used with Auto classes. It has no effect here and is ignored.
WARNING 06-03 19:28:58 [arg_utils.py:1175] `seed=None` is equivalent to `seed=0` in V1 Engine. You will no longer be allowed to pass `None` in v0.13.
INFO 06-03 19:28:58 [model.py:637] Resolved architecture: Qwen3ForCausalLM
WARNING 06-03 19:28:58 [model.py:2089] Casting torch.bfloat16 to torch.float16.
INFO 06-03 19:28:58 [model.py:1750] Using max model len 2048
INFO 06-03 19:28:58 [scheduler.py:228] Chunked prefill is enabled with max_num_batched_tokens=8192.
WARNING 06-03 19:28:59 [system_utils.py:136] We must use the `spawn` multiprocessing start method. Overriding VLLM_WORKER_MULTIPROC_METHOD to 'spawn'. See https://docs.vllm.ai/en/latest/usage/troubleshooting.html#python-multiprocessing for more information. Reasons: CUDA is initialized
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:07 [core.py:93] Initializing a V1 LLM engine (v0.12.0) with config: model='/root/.cache/modelscope/hub/models/FunAudioLLM/Fun-ASR-Nano-2512/Qwen3-0.6B-vllm', speculative_config=None, tokenizer='/root/.cache/modelscope/hub/models/FunAudioLLM/Fun-ASR-Nano-2512/Qwen3-0.6B-vllm', skip_tokenizer_init=False, tokenizer_mode=auto, revision=None, tokenizer_revision=None, trust_remote_code=True, dtype=torch.float16, max_seq_len=2048, download_dir=None, load_format=auto, tensor_parallel_size=1, pipeline_parallel_size=1, data_parallel_size=1, disable_custom_all_reduce=False, quantization=None, enforce_eager=False, kv_cache_dtype=auto, device_config=cuda, structured_outputs_config=StructuredOutputsConfig(backend='auto', disable_fallback=False, disable_any_whitespace=False, disable_additional_properties=False, reasoning_parser='', reasoning_parser_plugin='', enable_in_reasoning=False), observability_config=ObservabilityConfig(show_hidden_metrics_for_version=None, otlp_traces_endpoint=None, collect_detailed_traces=None, kv_cache_metrics=False, kv_cache_metrics_sample=0.01), seed=0, served_model_name=/root/.cache/modelscope/hub/models/FunAudioLLM/Fun-ASR-Nano-2512/Qwen3-0.6B-vllm, enable_prefix_caching=True, enable_chunked_prefill=True, pooler_config=None, compilation_config={'level': None, 'mode': <CompilationMode.VLLM_COMPILE: 3>, 'debug_dump_path': None, 'cache_dir': '', 'compile_cache_save_format': 'binary', 'backend': 'inductor', 'custom_ops': ['none'], 'splitting_ops': ['vllm::unified_attention', 'vllm::unified_attention_with_output', 'vllm::unified_mla_attention', 'vllm::unified_mla_attention_with_output', 'vllm::mamba_mixer2', 'vllm::mamba_mixer', 'vllm::short_conv', 'vllm::linear_attention', 'vllm::plamo2_mamba_mixer', 'vllm::gdn_attention_core', 'vllm::kda_attention', 'vllm::sparse_attn_indexer'], 'compile_mm_encoder': False, 'compile_sizes': [], 'inductor_compile_config': {'enable_auto_functionalized_v2': False, 'combo_kernels': True, 'benchmark_combo_kernel': True}, 'inductor_passes': {}, 'cudagraph_mode': <CUDAGraphMode.FULL_AND_PIECEWISE: (2, 1)>, 'cudagraph_num_of_warmups': 1, 'cudagraph_capture_sizes': [1, 2, 4, 8, 16, 24, 32, 40, 48, 56, 64, 72, 80, 88, 96, 104, 112, 120, 128, 136, 144, 152, 160, 168, 176, 184, 192, 200, 208, 216, 224, 232, 240, 248, 256, 272, 288, 304, 320, 336, 352, 368, 384, 400, 416, 432, 448, 464, 480, 496, 512], 'cudagraph_copy_inputs': False, 'cudagraph_specialize_lora': True, 'use_inductor_graph_partition': False, 'pass_config': {'fuse_norm_quant': False, 'fuse_act_quant': False, 'fuse_attn_quant': False, 'eliminate_noops': True, 'enable_sp': False, 'fuse_gemm_comms': False, 'fuse_allreduce_rms': False}, 'max_cudagraph_capture_size': 512, 'dynamic_shapes_config': {'type': <DynamicShapesType.BACKED: 'backed'>}, 'local_cache_dir': None}
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m ERROR 06-03 19:29:07 [fa_utils.py:72] Cannot use FA version 2 is not supported due to FA2 is only supported on devices with compute capability >= 8
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:07 [parallel_state.py:1200] world_size=1 rank=0 local_rank=0 distributed_init_method=tcp://10.125.210.20:60227 backend=nccl
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:08 [parallel_state.py:1408] rank 0 in world size 1 is assigned as DP rank 0, PP rank 0, PCP rank 0, TP rank 0, EP rank 0
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:08 [gpu_model_runner.py:3467] Starting to load model /root/.cache/modelscope/hub/models/FunAudioLLM/Fun-ASR-Nano-2512/Qwen3-0.6B-vllm...
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:09 [cuda.py:411] Using TRITON_ATTN attention backend out of potential backends: ['TRITON_ATTN', 'FLEX_ATTENTION']
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m 
Loading safetensors checkpoint shards:   0% Completed | 0/1 [00:00<?, ?it/s]
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m 
Loading safetensors checkpoint shards: 100% Completed | 1/1 [00:01<00:00,  1.29s/it]
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m 
Loading safetensors checkpoint shards: 100% Completed | 1/1 [00:01<00:00,  1.29s/it]
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m 
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:10 [default_loader.py:308] Loading weights took 1.39 seconds
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:11 [gpu_model_runner.py:3549] Model loading took 1.1201 GiB memory and 1.820741 seconds
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:20 [backends.py:655] Using cache directory: /root/.cache/vllm/torch_compile_cache/6cdaa4f620/rank_0_0/backbone for vLLM's torch.compile
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:20 [backends.py:715] Dynamo bytecode transform time: 8.98 s
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:24 [backends.py:216] Directly load the compiled graph(s) for dynamic shape from the cache, took 3.240 s
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:26 [monitor.py:34] torch.compile takes 12.22 s in total
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:27 [gpu_worker.py:359] Available KV cache memory: 13.31 GiB
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:27 [kv_cache_utils.py:1286] GPU KV cache size: 124,640 tokens
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:27 [kv_cache_utils.py:1291] Maximum concurrency for 2,048 tokens per request: 60.86x
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m 
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):   0%|          | 0/51 [00:00<?, ?it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):   4%|▍         | 2/51 [00:00<00:03, 14.97it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):   8%|▊         | 4/51 [00:00<00:03, 15.25it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  12%|█▏        | 6/51 [00:00<00:02, 16.54it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  16%|█▌        | 8/51 [00:00<00:02, 17.67it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  20%|█▉        | 10/51 [00:00<00:02, 15.34it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  24%|██▎       | 12/51 [00:00<00:02, 15.68it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  27%|██▋       | 14/51 [00:00<00:02, 16.15it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  33%|███▎      | 17/51 [00:01<00:01, 18.13it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  37%|███▋      | 19/51 [00:01<00:01, 18.14it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  41%|████      | 21/51 [00:01<00:01, 16.63it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  45%|████▌     | 23/51 [00:01<00:01, 15.13it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  51%|█████     | 26/51 [00:01<00:01, 17.21it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  57%|█████▋    | 29/51 [00:01<00:01, 18.72it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  61%|██████    | 31/51 [00:01<00:01, 17.36it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  65%|██████▍   | 33/51 [00:01<00:01, 16.86it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  69%|██████▊   | 35/51 [00:02<00:00, 16.56it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  75%|███████▍  | 38/51 [00:02<00:00, 18.27it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  80%|████████  | 41/51 [00:02<00:00, 18.33it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  84%|████████▍ | 43/51 [00:02<00:00, 16.38it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  88%|████████▊ | 45/51 [00:02<00:00, 16.40it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  92%|█████████▏| 47/51 [00:02<00:00, 17.07it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE):  98%|█████████▊| 50/51 [00:02<00:00, 18.32it/s]
Capturing CUDA graphs (mixed prefill-decode, PIECEWISE): 100%|██████████| 51/51 [00:03<00:00, 16.89it/s]
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m 
Capturing CUDA graphs (decode, FULL):   0%|          | 0/35 [00:00<?, ?it/s]
Capturing CUDA graphs (decode, FULL):   3%|▎         | 1/35 [00:00<00:04,  7.16it/s]
Capturing CUDA graphs (decode, FULL):   9%|▊         | 3/35 [00:00<00:02, 11.96it/s]
Capturing CUDA graphs (decode, FULL):  14%|█▍        | 5/35 [00:00<00:02, 14.82it/s]
Capturing CUDA graphs (decode, FULL):  20%|██        | 7/35 [00:00<00:01, 16.35it/s]
Capturing CUDA graphs (decode, FULL):  26%|██▌       | 9/35 [00:00<00:01, 14.82it/s]
Capturing CUDA graphs (decode, FULL):  31%|███▏      | 11/35 [00:00<00:01, 13.31it/s]
Capturing CUDA graphs (decode, FULL):  37%|███▋      | 13/35 [00:00<00:01, 14.79it/s]
Capturing CUDA graphs (decode, FULL):  43%|████▎     | 15/35 [00:01<00:01, 16.10it/s]
Capturing CUDA graphs (decode, FULL):  49%|████▊     | 17/35 [00:01<00:01, 15.35it/s]
Capturing CUDA graphs (decode, FULL):  54%|█████▍    | 19/35 [00:01<00:01, 13.73it/s]
Capturing CUDA graphs (decode, FULL):  60%|██████    | 21/35 [00:01<00:00, 14.04it/s]
Capturing CUDA graphs (decode, FULL):  66%|██████▌   | 23/35 [00:01<00:00, 15.39it/s]
Capturing CUDA graphs (decode, FULL):  71%|███████▏  | 25/35 [00:01<00:00, 16.38it/s]
Capturing CUDA graphs (decode, FULL):  77%|███████▋  | 27/35 [00:01<00:00, 14.90it/s]
Capturing CUDA graphs (decode, FULL):  83%|████████▎ | 29/35 [00:01<00:00, 15.08it/s]
Capturing CUDA graphs (decode, FULL):  89%|████████▊ | 31/35 [00:02<00:00, 15.27it/s]
Capturing CUDA graphs (decode, FULL):  94%|█████████▍| 33/35 [00:02<00:00, 15.46it/s]
Capturing CUDA graphs (decode, FULL): 100%|██████████| 35/35 [00:02<00:00, 14.73it/s]
Capturing CUDA graphs (decode, FULL): 100%|██████████| 35/35 [00:02<00:00, 14.72it/s]
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:33 [gpu_model_runner.py:4466] Graph capturing finished in 6 secs, took 0.50 GiB
�[0;36m(EngineCore_DP0 pid=2084414)�[0;0m INFO 06-03 19:29:33 [core.py:254] init engine (profile, create kv cache, warmup model) took 22.66 seconds
INFO 06-03 19:29:35 [llm.py:343] Supported tasks: ['generate']
[rank0]:[W603 19:29:39.588427385 ProcessGroupNCCL.cpp:1524] Warning: WARNING: destroy_process_group() was not called before program exit, which can leak resources. For more info, please see https://pytorch.org/docs/stable/distributed.html#shutdown (function operator())
Model loaded in 49.6s


Processing 1 audio file(s)...
  Batch 1/1: 1 files in 2.92s

============================================================
Results: 1 samples, total inference time: 2.92s
============================================================

[zh]
  Text: 

</details> ```