Profile Job Results
Jobs
jlpenv61p
Results Ready
Name
whisper_base_en_WhisperEncoder
Target Device
- SA8650 (Proxy)
- Android 13
- Qualcomm® SA8650P
Creator
ai-hub-support@qti.qualcomm.com
Target Model
Input Specs
audio: float32[1, 80, 3000]Completion Time
8/27/2024, 8:24:36 AM
Options
--compute_unit gpu --max_profiler_iterations 10Versions
- TensorFlow Lite: 2.16.1
- Android: 13 (TP1A.220624.014)
- AI Hub: aihub-2024.08.23.0
Estimated Inference Time
160 ms
Estimated Peak Memory Usage
23 - 124 MB
Compute Units
GPU
419
| Stage | Time | Memory |
|---|---|---|
First App Load | 5.00 s | 310-312 MB |
Subsequent App Load | 328 ms | 115-194 MB |
Inference | 160 ms | 23-124 MB |
| TensorFlow Lite | Value |
|---|---|
| number_of_threads | 4 |
| GPUv2 Delegate | Value |
|---|---|
| inference_preference | TFLITE_GPU_INFERENCE_PREFERENCE_SUSTAINED_SPEED |
| inference_priority1 | TFLITE_GPU_INFERENCE_PRIORITY_MIN_LATENCY |
| inference_priority2 | TFLITE_GPU_INFERENCE_PRIORITY_MIN_MEMORY_USAGE |
| inference_priority3 | TFLITE_GPU_INFERENCE_PRIORITY_MAX_PRECISION |
Sign up to run this model on a hosted Qualcomm® device!
Run on device