qaihm-bot commited on
Commit
fa4927a
·
verified ·
1 Parent(s): a9541b6

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +71 -72
README.md CHANGED
@@ -1,6 +1,6 @@
1
  ---
2
  library_name: pytorch
3
- license: mit
4
  tags:
5
  - foundation
6
  - android
@@ -26,7 +26,7 @@ More details on model performance across various devices, can be found
26
 
27
  ### Model Details
28
 
29
- - **Model Type:** Speech recognition
30
  - **Model Stats:**
31
  - Model checkpoint: base.en
32
  - Input resolution: 80x3000 (30 seconds audio)
@@ -36,65 +36,64 @@ More details on model performance across various devices, can be found
36
  - Number of parameters (WhisperDecoder): 48.6M
37
  - Model size (WhisperDecoder): 186 MB
38
 
39
- | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
40
  |---|---|---|---|---|---|---|---|---|
41
- | WhisperEncoderInf | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 207.168 ms | 0 - 67 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.tflite) |
42
- | WhisperEncoderInf | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 270.321 ms | 0 - 355 MB | FP16 | NPU | [Whisper-Base-En.so](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.so) |
43
- | WhisperEncoderInf | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 258.293 ms | 53 - 564 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.onnx) |
44
- | WhisperEncoderInf | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 159.338 ms | 39 - 85 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.tflite) |
45
- | WhisperEncoderInf | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 225.666 ms | 0 - 1377 MB | FP16 | NPU | [Whisper-Base-En.so](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.so) |
46
- | WhisperEncoderInf | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 165.997 ms | 163 - 1665 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.onnx) |
47
- | WhisperEncoderInf | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 129.548 ms | 39 - 67 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.tflite) |
48
- | WhisperEncoderInf | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 159.839 ms | 41 - 1544 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.onnx) |
49
- | WhisperEncoderInf | SA7255P ADP | SA7255P | TFLITE | 1153.482 ms | 37 - 60 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.tflite) |
50
- | WhisperEncoderInf | SA7255P ADP | SA7255P | QNN | 1010.596 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
51
- | WhisperEncoderInf | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 204.974 ms | 0 - 67 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.tflite) |
52
- | WhisperEncoderInf | SA8255 (Proxy) | SA8255P Proxy | QNN | 251.25 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
53
- | WhisperEncoderInf | SA8295P ADP | SA8295P | TFLITE | 205.071 ms | 38 - 69 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.tflite) |
54
- | WhisperEncoderInf | SA8295P ADP | SA8295P | QNN | 220.65 ms | 1 - 17 MB | FP16 | NPU | Use Export Script |
55
- | WhisperEncoderInf | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 233.593 ms | 0 - 77 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.tflite) |
56
- | WhisperEncoderInf | SA8650 (Proxy) | SA8650P Proxy | QNN | 221.139 ms | 1 - 2 MB | FP16 | NPU | Use Export Script |
57
- | WhisperEncoderInf | SA8775P ADP | SA8775P | TFLITE | 367.2 ms | 38 - 62 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.tflite) |
58
- | WhisperEncoderInf | SA8775P ADP | SA8775P | QNN | 215.878 ms | 0 - 9 MB | FP16 | NPU | Use Export Script |
59
- | WhisperEncoderInf | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 1153.482 ms | 37 - 60 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.tflite) |
60
- | WhisperEncoderInf | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 1010.596 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
61
- | WhisperEncoderInf | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 201.097 ms | 0 - 66 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.tflite) |
62
- | WhisperEncoderInf | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 242.356 ms | 1 - 4 MB | FP16 | NPU | Use Export Script |
63
- | WhisperEncoderInf | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 367.2 ms | 38 - 62 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.tflite) |
64
- | WhisperEncoderInf | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 215.878 ms | 0 - 9 MB | FP16 | NPU | Use Export Script |
65
- | WhisperEncoderInf | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 264.328 ms | 38 - 89 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.tflite) |
66
- | WhisperEncoderInf | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 383.819 ms | 1 - 1441 MB | FP16 | NPU | Use Export Script |
67
- | WhisperEncoderInf | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 174.12 ms | 0 - 0 MB | FP16 | NPU | Use Export Script |
68
- | WhisperEncoderInf | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 202.131 ms | 133 - 133 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoderInf.onnx) |
69
- | WhisperDecoderInf | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 9.835 ms | 5 - 31 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.tflite) |
70
- | WhisperDecoderInf | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 5.8 ms | 20 - 39 MB | FP16 | NPU | [Whisper-Base-En.so](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.so) |
71
- | WhisperDecoderInf | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 10.337 ms | 11 - 446 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.onnx) |
72
- | WhisperDecoderInf | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 7.583 ms | 5 - 123 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.tflite) |
73
- | WhisperDecoderInf | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 4.397 ms | 186 - 258 MB | FP16 | NPU | [Whisper-Base-En.so](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.so) |
74
- | WhisperDecoderInf | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 8.155 ms | 50 - 178 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.onnx) |
75
- | WhisperDecoderInf | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 7.295 ms | 5 - 114 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.tflite) |
76
- | WhisperDecoderInf | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 3.486 ms | 18 - 84 MB | FP16 | NPU | Use Export Script |
77
- | WhisperDecoderInf | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 7.88 ms | 50 - 155 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.onnx) |
78
- | WhisperDecoderInf | SA7255P ADP | SA7255P | TFLITE | 36.508 ms | 5 - 111 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.tflite) |
79
- | WhisperDecoderInf | SA7255P ADP | SA7255P | QNN | 26.501 ms | 18 - 27 MB | FP16 | NPU | Use Export Script |
80
- | WhisperDecoderInf | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 9.908 ms | 5 - 31 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.tflite) |
81
- | WhisperDecoderInf | SA8255 (Proxy) | SA8255P Proxy | QNN | 4.359 ms | 20 - 22 MB | FP16 | NPU | Use Export Script |
82
- | WhisperDecoderInf | SA8295P ADP | SA8295P | TFLITE | 12.313 ms | 5 - 104 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.tflite) |
83
- | WhisperDecoderInf | SA8295P ADP | SA8295P | QNN | 5.765 ms | 18 - 35 MB | FP16 | NPU | Use Export Script |
84
- | WhisperDecoderInf | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 10.0 ms | 5 - 28 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.tflite) |
85
- | WhisperDecoderInf | SA8650 (Proxy) | SA8650P Proxy | QNN | 4.213 ms | 20 - 23 MB | FP16 | NPU | Use Export Script |
86
- | WhisperDecoderInf | SA8775P ADP | SA8775P | TFLITE | 12.335 ms | 4 - 109 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.tflite) |
87
- | WhisperDecoderInf | SA8775P ADP | SA8775P | QNN | 5.42 ms | 18 - 27 MB | FP16 | NPU | Use Export Script |
88
- | WhisperDecoderInf | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 36.508 ms | 5 - 111 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.tflite) |
89
- | WhisperDecoderInf | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 26.501 ms | 18 - 27 MB | FP16 | NPU | Use Export Script |
90
- | WhisperDecoderInf | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 9.84 ms | 6 - 31 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.tflite) |
91
- | WhisperDecoderInf | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 4.237 ms | 20 - 24 MB | FP16 | NPU | Use Export Script |
92
- | WhisperDecoderInf | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 12.335 ms | 4 - 109 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.tflite) |
93
- | WhisperDecoderInf | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 5.42 ms | 18 - 27 MB | FP16 | NPU | Use Export Script |
94
- | WhisperDecoderInf | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 12.029 ms | 5 - 116 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.tflite) |
95
- | WhisperDecoderInf | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 6.666 ms | 20 - 91 MB | FP16 | NPU | Use Export Script |
96
- | WhisperDecoderInf | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 3.797 ms | 20 - 20 MB | FP16 | NPU | Use Export Script |
97
- | WhisperDecoderInf | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 9.212 ms | 106 - 106 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoderInf.onnx) |
98
 
99
 
100
 
@@ -156,21 +155,21 @@ python -m qai_hub_models.models.whisper_base_en.export
156
  Profiling Results
157
  ------------------------------------------------------------
158
  WhisperEncoderInf
159
- Device : Samsung Galaxy S23 (13)
160
- Runtime : TFLITE
161
- Estimated inference time (ms) : 207.2
162
- Estimated peak memory usage (MB): [0, 67]
163
- Total # Ops : 419
164
- Compute Unit(s) : GPU (408 ops) CPU (11 ops)
165
 
166
  ------------------------------------------------------------
167
  WhisperDecoderInf
168
- Device : Samsung Galaxy S23 (13)
169
- Runtime : TFLITE
170
- Estimated inference time (ms) : 9.8
171
- Estimated peak memory usage (MB): [5, 31]
172
- Total # Ops : 983
173
- Compute Unit(s) : NPU (983 ops)
174
  ```
175
 
176
 
 
1
  ---
2
  library_name: pytorch
3
+ license: other
4
  tags:
5
  - foundation
6
  - android
 
26
 
27
  ### Model Details
28
 
29
+ - **Model Type:** Model_use_case.speech_recognition
30
  - **Model Stats:**
31
  - Model checkpoint: base.en
32
  - Input resolution: 80x3000 (30 seconds audio)
 
36
  - Number of parameters (WhisperDecoder): 48.6M
37
  - Model size (WhisperDecoder): 186 MB
38
 
39
+ | Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
40
  |---|---|---|---|---|---|---|---|---|
41
+ | WhisperEncoderInf | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 862.37 ms | 37 - 59 MB | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
42
+ | WhisperEncoderInf | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 350.075 ms | 0 - 10 MB | NPU | Use Export Script |
43
+ | WhisperEncoderInf | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 263.156 ms | 38 - 87 MB | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
44
+ | WhisperEncoderInf | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 394.245 ms | 0 - 1424 MB | NPU | Use Export Script |
45
+ | WhisperEncoderInf | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 204.387 ms | 0 - 68 MB | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
46
+ | WhisperEncoderInf | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 245.146 ms | 1 - 3 MB | NPU | Use Export Script |
47
+ | WhisperEncoderInf | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 353.075 ms | 38 - 62 MB | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
48
+ | WhisperEncoderInf | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 188.7 ms | 1 - 11 MB | NPU | Use Export Script |
49
+ | WhisperEncoderInf | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 862.37 ms | 37 - 59 MB | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
50
+ | WhisperEncoderInf | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 350.075 ms | 0 - 10 MB | NPU | Use Export Script |
51
+ | WhisperEncoderInf | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 200.902 ms | 0 - 61 MB | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
52
+ | WhisperEncoderInf | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 216.589 ms | 1 - 3 MB | NPU | Use Export Script |
53
+ | WhisperEncoderInf | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 191.412 ms | 37 - 68 MB | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
54
+ | WhisperEncoderInf | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 220.687 ms | 1 - 17 MB | NPU | Use Export Script |
55
+ | WhisperEncoderInf | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 207.467 ms | 0 - 70 MB | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
56
+ | WhisperEncoderInf | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 246.912 ms | 1 - 3 MB | NPU | Use Export Script |
57
+ | WhisperEncoderInf | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 353.075 ms | 38 - 62 MB | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
58
+ | WhisperEncoderInf | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 188.7 ms | 1 - 11 MB | NPU | Use Export Script |
59
+ | WhisperEncoderInf | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 205.105 ms | 0 - 63 MB | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
60
+ | WhisperEncoderInf | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 189.365 ms | 0 - 356 MB | NPU | Use Export Script |
61
+ | WhisperEncoderInf | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 190.475 ms | 53 - 566 MB | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.onnx) |
62
+ | WhisperEncoderInf | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 154.58 ms | 37 - 80 MB | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
63
+ | WhisperEncoderInf | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 194.574 ms | 0 - 1371 MB | NPU | Use Export Script |
64
+ | WhisperEncoderInf | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 166.085 ms | 71 - 1577 MB | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.onnx) |
65
+ | WhisperEncoderInf | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 157.782 ms | 37 - 65 MB | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
66
+ | WhisperEncoderInf | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 145.061 ms | 67 - 1573 MB | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.onnx) |
67
+ | WhisperEncoderInf | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 173.97 ms | 1 - 1 MB | NPU | Use Export Script |
68
+ | WhisperEncoderInf | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 203.45 ms | 131 - 131 MB | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.onnx) |
69
+ | WhisperDecoderInf | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | TFLITE | 17.035 ms | 6 - 111 MB | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
70
+ | WhisperDecoderInf | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN | 6.869 ms | 15 - 24 MB | NPU | Use Export Script |
71
+ | WhisperDecoderInf | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | TFLITE | 13.376 ms | 5 - 115 MB | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
72
+ | WhisperDecoderInf | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN | 8.599 ms | 20 - 89 MB | NPU | Use Export Script |
73
+ | WhisperDecoderInf | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | TFLITE | 9.902 ms | 0 - 27 MB | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
74
+ | WhisperDecoderInf | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN | 4.194 ms | 20 - 23 MB | NPU | Use Export Script |
75
+ | WhisperDecoderInf | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | TFLITE | 11.172 ms | 3 - 109 MB | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
76
+ | WhisperDecoderInf | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN | 5.135 ms | 15 - 25 MB | NPU | Use Export Script |
77
+ | WhisperDecoderInf | float | SA7255P ADP | Qualcomm® SA7255P | TFLITE | 17.035 ms | 6 - 111 MB | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
78
+ | WhisperDecoderInf | float | SA7255P ADP | Qualcomm® SA7255P | QNN | 6.869 ms | 15 - 24 MB | NPU | Use Export Script |
79
+ | WhisperDecoderInf | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | TFLITE | 9.835 ms | 0 - 25 MB | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
80
+ | WhisperDecoderInf | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN | 4.335 ms | 20 - 22 MB | NPU | Use Export Script |
81
+ | WhisperDecoderInf | float | SA8295P ADP | Qualcomm® SA8295P | TFLITE | 12.218 ms | 5 - 104 MB | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
82
+ | WhisperDecoderInf | float | SA8295P ADP | Qualcomm® SA8295P | QNN | 5.66 ms | 18 - 35 MB | NPU | Use Export Script |
83
+ | WhisperDecoderInf | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | TFLITE | 9.857 ms | 5 - 30 MB | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
84
+ | WhisperDecoderInf | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN | 4.265 ms | 21 - 24 MB | NPU | Use Export Script |
85
+ | WhisperDecoderInf | float | SA8775P ADP | Qualcomm® SA8775P | TFLITE | 11.172 ms | 3 - 109 MB | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
86
+ | WhisperDecoderInf | float | SA8775P ADP | Qualcomm® SA8775P | QNN | 5.135 ms | 15 - 25 MB | NPU | Use Export Script |
87
+ | WhisperDecoderInf | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | TFLITE | 9.862 ms | 5 - 26 MB | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
88
+ | WhisperDecoderInf | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | QNN | 4.233 ms | 20 - 46 MB | NPU | Use Export Script |
89
+ | WhisperDecoderInf | float | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 Mobile | ONNX | 10.289 ms | 11 - 374 MB | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.onnx) |
90
+ | WhisperDecoderInf | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | TFLITE | 7.572 ms | 4 - 120 MB | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
91
+ | WhisperDecoderInf | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN | 3.273 ms | 20 - 81 MB | NPU | Use Export Script |
92
+ | WhisperDecoderInf | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | ONNX | 7.681 ms | 50 - 170 MB | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.onnx) |
93
+ | WhisperDecoderInf | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | TFLITE | 7.283 ms | 5 - 115 MB | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.tflite) |
94
+ | WhisperDecoderInf | float | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite Mobile | ONNX | 7.764 ms | 52 - 158 MB | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.onnx) |
95
+ | WhisperDecoderInf | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 3.798 ms | 20 - 20 MB | NPU | Use Export Script |
96
+ | WhisperDecoderInf | float | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 9.088 ms | 106 - 106 MB | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/Whisper-Base-En.onnx) |
 
97
 
98
 
99
 
 
155
  Profiling Results
156
  ------------------------------------------------------------
157
  WhisperEncoderInf
158
+ Device : cs_8275 (ANDROID 14)
159
+ Runtime : TFLITE
160
+ Estimated inference time (ms) : 862.4
161
+ Estimated peak memory usage (MB): [37, 59]
162
+ Total # Ops : 419
163
+ Compute Unit(s) : npu (0 ops) gpu (408 ops) cpu (11 ops)
164
 
165
  ------------------------------------------------------------
166
  WhisperDecoderInf
167
+ Device : cs_8275 (ANDROID 14)
168
+ Runtime : TFLITE
169
+ Estimated inference time (ms) : 17.0
170
+ Estimated peak memory usage (MB): [6, 111]
171
+ Total # Ops : 983
172
+ Compute Unit(s) : npu (983 ops) gpu (0 ops) cpu (0 ops)
173
  ```
174
 
175