whisperx-small-npu / config.json
magicunicorn's picture
Upload whisperx-small-npu - INT8 quantized for AMD NPU
99b80e0 verified
{
"architectures": [
"WhisperForConditionalGeneration"
],
"model_type": "whisper",
"quantization": {
"method": "INT8",
"backend": "NPU-AIE2",
"hardware": "AMD Phoenix NPU",
"performance_rtf": "0.003 RTF",
"tokens_per_second": 4789
},
"npu_config": {
"tiles": 20,
"vector_width": 32,
"dma_channels": 2,
"kernel_type": "MLIR-AIE2",
"optimization_level": 3
},
"audio": {
"sampling_rate": 16000,
"chunk_length": 30,
"n_mels": 80
},
"base_model": "openai/whisper-small",
"implementation": "unicorn-engine",
"license": "mit"
}