Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
@@ -38,56 +38,56 @@ More details on model performance across various devices, can be found
|
|
38 |
|
39 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
40 |
|---|---|---|---|---|---|---|---|---|
|
41 |
-
| WhisperDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE |
|
42 |
-
| WhisperDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 4.
|
43 |
-
| WhisperDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX |
|
44 |
-
| WhisperDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 32.
|
45 |
-
| WhisperDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 3.
|
46 |
-
| WhisperDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 13.
|
47 |
-
| WhisperDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE |
|
48 |
-
| WhisperDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 2.
|
49 |
-
| WhisperDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 12.
|
50 |
-
| WhisperDecoder | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 38.
|
51 |
-
| WhisperDecoder | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 4.
|
52 |
-
| WhisperDecoder | SA7255P ADP | SA7255P | TFLITE |
|
53 |
-
| WhisperDecoder | SA7255P ADP | SA7255P | QNN | 26.
|
54 |
-
| WhisperDecoder | SA8255 (Proxy) | SA8255P Proxy | TFLITE |
|
55 |
-
| WhisperDecoder | SA8255 (Proxy) | SA8255P Proxy | QNN | 4.
|
56 |
-
| WhisperDecoder | SA8295P ADP | SA8295P | TFLITE |
|
57 |
-
| WhisperDecoder | SA8295P ADP | SA8295P | QNN | 5.
|
58 |
-
| WhisperDecoder | SA8650 (Proxy) | SA8650P Proxy | TFLITE |
|
59 |
-
| WhisperDecoder | SA8650 (Proxy) | SA8650P Proxy | QNN | 4.
|
60 |
-
| WhisperDecoder | SA8775P ADP | SA8775P | TFLITE |
|
61 |
-
| WhisperDecoder | SA8775P ADP | SA8775P | QNN | 5.
|
62 |
-
| WhisperDecoder | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 42.
|
63 |
-
| WhisperDecoder | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 5.
|
64 |
-
| WhisperDecoder | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 3.
|
65 |
-
| WhisperDecoder | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 14.
|
66 |
-
| WhisperEncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE |
|
67 |
-
| WhisperEncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN |
|
68 |
-
| WhisperEncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX |
|
69 |
-
| WhisperEncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE |
|
70 |
-
| WhisperEncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN |
|
71 |
-
| WhisperEncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX |
|
72 |
-
| WhisperEncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE |
|
73 |
-
| WhisperEncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 152.
|
74 |
-
| WhisperEncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX |
|
75 |
-
| WhisperEncoder | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE |
|
76 |
-
| WhisperEncoder | QCS8550 (Proxy) | QCS8550 Proxy | QNN |
|
77 |
-
| WhisperEncoder | SA7255P ADP | SA7255P | TFLITE |
|
78 |
-
| WhisperEncoder | SA7255P ADP | SA7255P | QNN |
|
79 |
-
| WhisperEncoder | SA8255 (Proxy) | SA8255P Proxy | TFLITE |
|
80 |
-
| WhisperEncoder | SA8255 (Proxy) | SA8255P Proxy | QNN |
|
81 |
-
| WhisperEncoder | SA8295P ADP | SA8295P | TFLITE | 204.
|
82 |
-
| WhisperEncoder | SA8295P ADP | SA8295P | QNN | 219.
|
83 |
-
| WhisperEncoder | SA8650 (Proxy) | SA8650P Proxy | TFLITE |
|
84 |
-
| WhisperEncoder | SA8650 (Proxy) | SA8650P Proxy | QNN |
|
85 |
-
| WhisperEncoder | SA8775P ADP | SA8775P | TFLITE |
|
86 |
-
| WhisperEncoder | SA8775P ADP | SA8775P | QNN |
|
87 |
-
| WhisperEncoder | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE |
|
88 |
-
| WhisperEncoder | QCS8450 (Proxy) | QCS8450 Proxy | QNN |
|
89 |
-
| WhisperEncoder | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 160.
|
90 |
-
| WhisperEncoder | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 297.
|
91 |
|
92 |
|
93 |
|
@@ -152,8 +152,8 @@ Profiling Results
|
|
152 |
WhisperDecoder
|
153 |
Device : Samsung Galaxy S23 (13)
|
154 |
Runtime : TFLITE
|
155 |
-
Estimated inference time (ms) :
|
156 |
-
Estimated peak memory usage (MB): [
|
157 |
Total # Ops : 983
|
158 |
Compute Unit(s) : NPU (983 ops)
|
159 |
|
@@ -161,8 +161,8 @@ Compute Unit(s) : NPU (983 ops)
|
|
161 |
WhisperEncoder
|
162 |
Device : Samsung Galaxy S23 (13)
|
163 |
Runtime : TFLITE
|
164 |
-
Estimated inference time (ms) :
|
165 |
-
Estimated peak memory usage (MB): [
|
166 |
Total # Ops : 419
|
167 |
Compute Unit(s) : GPU (408 ops) CPU (11 ops)
|
168 |
```
|
|
|
38 |
|
39 |
| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|
40 |
|---|---|---|---|---|---|---|---|---|
|
41 |
+
| WhisperDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 38.913 ms | 2 - 40 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.tflite) |
|
42 |
+
| WhisperDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 4.2 ms | 20 - 63 MB | FP16 | NPU | [Whisper-Base-En.so](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.so) |
|
43 |
+
| WhisperDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 16.043 ms | 17 - 204 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.onnx) |
|
44 |
+
| WhisperDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 32.737 ms | 5 - 85 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.tflite) |
|
45 |
+
| WhisperDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 3.241 ms | 21 - 75 MB | FP16 | NPU | [Whisper-Base-En.so](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.so) |
|
46 |
+
| WhisperDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 13.836 ms | 48 - 123 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.onnx) |
|
47 |
+
| WhisperDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 28.728 ms | 4 - 77 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.tflite) |
|
48 |
+
| WhisperDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 2.575 ms | 9 - 61 MB | FP16 | NPU | Use Export Script |
|
49 |
+
| WhisperDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 12.41 ms | 52 - 123 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.onnx) |
|
50 |
+
| WhisperDecoder | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 38.898 ms | 5 - 41 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.tflite) |
|
51 |
+
| WhisperDecoder | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 4.204 ms | 20 - 23 MB | FP16 | NPU | Use Export Script |
|
52 |
+
| WhisperDecoder | SA7255P ADP | SA7255P | TFLITE | 69.303 ms | 6 - 77 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.tflite) |
|
53 |
+
| WhisperDecoder | SA7255P ADP | SA7255P | QNN | 26.928 ms | 18 - 27 MB | FP16 | NPU | Use Export Script |
|
54 |
+
| WhisperDecoder | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 39.507 ms | 4 - 40 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.tflite) |
|
55 |
+
| WhisperDecoder | SA8255 (Proxy) | SA8255P Proxy | QNN | 4.275 ms | 19 - 21 MB | FP16 | NPU | Use Export Script |
|
56 |
+
| WhisperDecoder | SA8295P ADP | SA8295P | TFLITE | 41.075 ms | 6 - 73 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.tflite) |
|
57 |
+
| WhisperDecoder | SA8295P ADP | SA8295P | QNN | 5.685 ms | 18 - 32 MB | FP16 | NPU | Use Export Script |
|
58 |
+
| WhisperDecoder | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 39.484 ms | 5 - 41 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.tflite) |
|
59 |
+
| WhisperDecoder | SA8650 (Proxy) | SA8650P Proxy | QNN | 4.293 ms | 20 - 22 MB | FP16 | NPU | Use Export Script |
|
60 |
+
| WhisperDecoder | SA8775P ADP | SA8775P | TFLITE | 39.427 ms | 6 - 77 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.tflite) |
|
61 |
+
| WhisperDecoder | SA8775P ADP | SA8775P | QNN | 5.544 ms | 18 - 27 MB | FP16 | NPU | Use Export Script |
|
62 |
+
| WhisperDecoder | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 42.323 ms | 0 - 76 MB | FP16 | NPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.tflite) |
|
63 |
+
| WhisperDecoder | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 5.017 ms | 20 - 75 MB | FP16 | NPU | Use Export Script |
|
64 |
+
| WhisperDecoder | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 3.852 ms | 20 - 20 MB | FP16 | NPU | Use Export Script |
|
65 |
+
| WhisperDecoder | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 14.43 ms | 108 - 108 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperDecoder.onnx) |
|
66 |
+
| WhisperEncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 337.754 ms | 0 - 82 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.tflite) |
|
67 |
+
| WhisperEncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 273.538 ms | 0 - 84 MB | FP16 | NPU | [Whisper-Base-En.so](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.so) |
|
68 |
+
| WhisperEncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 281.449 ms | 71 - 273 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.onnx) |
|
69 |
+
| WhisperEncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 159.488 ms | 37 - 81 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.tflite) |
|
70 |
+
| WhisperEncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 185.633 ms | 13 - 300 MB | FP16 | NPU | [Whisper-Base-En.so](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.so) |
|
71 |
+
| WhisperEncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 205.979 ms | 66 - 395 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.onnx) |
|
72 |
+
| WhisperEncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 150.003 ms | 39 - 64 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.tflite) |
|
73 |
+
| WhisperEncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 152.344 ms | 0 - 304 MB | FP16 | NPU | Use Export Script |
|
74 |
+
| WhisperEncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 164.022 ms | 71 - 403 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.onnx) |
|
75 |
+
| WhisperEncoder | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 237.381 ms | 0 - 61 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.tflite) |
|
76 |
+
| WhisperEncoder | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 220.272 ms | 1 - 2 MB | FP16 | NPU | Use Export Script |
|
77 |
+
| WhisperEncoder | SA7255P ADP | SA7255P | TFLITE | 1155.427 ms | 37 - 59 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.tflite) |
|
78 |
+
| WhisperEncoder | SA7255P ADP | SA7255P | QNN | 937.355 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
|
79 |
+
| WhisperEncoder | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 236.135 ms | 0 - 64 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.tflite) |
|
80 |
+
| WhisperEncoder | SA8255 (Proxy) | SA8255P Proxy | QNN | 209.326 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
|
81 |
+
| WhisperEncoder | SA8295P ADP | SA8295P | TFLITE | 204.514 ms | 38 - 69 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.tflite) |
|
82 |
+
| WhisperEncoder | SA8295P ADP | SA8295P | QNN | 219.035 ms | 1 - 15 MB | FP16 | NPU | Use Export Script |
|
83 |
+
| WhisperEncoder | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 203.287 ms | 0 - 56 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.tflite) |
|
84 |
+
| WhisperEncoder | SA8650 (Proxy) | SA8650P Proxy | QNN | 208.865 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
|
85 |
+
| WhisperEncoder | SA8775P ADP | SA8775P | TFLITE | 365.818 ms | 38 - 63 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.tflite) |
|
86 |
+
| WhisperEncoder | SA8775P ADP | SA8775P | QNN | 194.677 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
|
87 |
+
| WhisperEncoder | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 268.324 ms | 38 - 88 MB | FP16 | GPU | [Whisper-Base-En.tflite](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.tflite) |
|
88 |
+
| WhisperEncoder | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 322.888 ms | 1 - 298 MB | FP16 | NPU | Use Export Script |
|
89 |
+
| WhisperEncoder | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 160.084 ms | 0 - 0 MB | FP16 | NPU | Use Export Script |
|
90 |
+
| WhisperEncoder | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 297.755 ms | 133 - 133 MB | FP16 | NPU | [Whisper-Base-En.onnx](https://huggingface.co/qualcomm/Whisper-Base-En/blob/main/WhisperEncoder.onnx) |
|
91 |
|
92 |
|
93 |
|
|
|
152 |
WhisperDecoder
|
153 |
Device : Samsung Galaxy S23 (13)
|
154 |
Runtime : TFLITE
|
155 |
+
Estimated inference time (ms) : 38.9
|
156 |
+
Estimated peak memory usage (MB): [2, 40]
|
157 |
Total # Ops : 983
|
158 |
Compute Unit(s) : NPU (983 ops)
|
159 |
|
|
|
161 |
WhisperEncoder
|
162 |
Device : Samsung Galaxy S23 (13)
|
163 |
Runtime : TFLITE
|
164 |
+
Estimated inference time (ms) : 337.8
|
165 |
+
Estimated peak memory usage (MB): [0, 82]
|
166 |
Total # Ops : 419
|
167 |
Compute Unit(s) : GPU (408 ops) CPU (11 ops)
|
168 |
```
|