Whisper-Base: Optimized for Qualcomm Devices
HuggingFace Whisper-Small ASR (Automatic Speech Recognition) model is a state-of-the-art system designed for transcribing spoken language into written text. This model is based on the transformer architecture and has been optimized for edge inference by replacing Multi-Head Attention (MHA) with Single-Head Attention (SHA) and linear layers with convolutional (conv) layers. It exhibits robust performance in realistic, noisy environments, making it highly reliable for real-world applications. Specifically, it excels in long-form transcription, capable of accurately transcribing audio clips up to 30 seconds long. Time to the first token is the encoder's latency, while time to each additional token is decoder's latency, where we assume a max decoded length specified below.
This is based on the implementation of Whisper-Base found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.
Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.
Getting Started
There are two ways to deploy this model on your device:
Option 1: Download Pre-Exported Models
Below are pre-exported model assets ready for deployment.
| Runtime | Precision | Chipset | SDK Versions | Download |
|---|---|---|---|---|
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | QAIRT 2.42, ONNX Runtime 1.24.1 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | QAIRT 2.43 | Download |
| QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | QAIRT 2.43 | Download |
For more device-specific assets and performance metrics, visit Whisper-Base on Qualcomm® AI Hub.
Option 2: Export with Custom Configurations
Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:
- Custom weights (e.g., fine-tuned checkpoints)
- Custom input shapes
- Target device and runtime configurations
This option is ideal if you need to customize the model beyond the default configuration provided here.
See our repository for Whisper-Base on GitHub for usage instructions.
Model Details
Model Type: Model_use_case.speech_recognition
Model Stats:
- Model checkpoint: openai/whisper-base
- Input resolution: 80x3000 (30 seconds audio)
- Max decoded sequence length: 200 tokens
- Number of parameters (HfWhisperEncoder): 23.7M
- Model size (HfWhisperEncoder) (float): 90.7 MB
- Number of parameters (HfWhisperDecoder): 48.9M
- Model size (HfWhisperDecoder) (float): 187 MB
Performance Summary
| Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |
|---|---|---|---|---|---|---|
| HfWhisperDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.476 ms | 20 - 30 MB | NPU |
| HfWhisperDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 1.896 ms | 126 - 126 MB | NPU |
| HfWhisperDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 3.296 ms | 125 - 125 MB | NPU |
| HfWhisperDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 3.21 ms | 0 - 8 MB | NPU |
| HfWhisperDecoder | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | 4.013 ms | 20 - 22 MB | NPU |
| HfWhisperDecoder | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | 4.629 ms | 20 - 43 MB | NPU |
| HfWhisperDecoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 2.692 ms | 3 - 10 MB | NPU |
| HfWhisperDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 2.473 ms | 20 - 30 MB | NPU |
| HfWhisperDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 2.416 ms | 20 - 20 MB | NPU |
| HfWhisperDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 3.327 ms | 20 - 20 MB | NPU |
| HfWhisperDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 3.195 ms | 20 - 27 MB | NPU |
| HfWhisperDecoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 6.533 ms | 20 - 28 MB | NPU |
| HfWhisperDecoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | 3.873 ms | 17 - 19 MB | NPU |
| HfWhisperDecoder | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | 4.864 ms | 20 - 29 MB | NPU |
| HfWhisperDecoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | 4.663 ms | 20 - 44 MB | NPU |
| HfWhisperDecoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | 5.161 ms | 20 - 28 MB | NPU |
| HfWhisperDecoder | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | 6.533 ms | 20 - 28 MB | NPU |
| HfWhisperDecoder | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | 5.414 ms | 18 - 23 MB | NPU |
| HfWhisperDecoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 2.648 ms | 0 - 9 MB | NPU |
| HfWhisperEncoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 14.926 ms | 38 - 48 MB | NPU |
| HfWhisperEncoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 15.4 ms | 64 - 64 MB | NPU |
| HfWhisperEncoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 37.311 ms | 67 - 67 MB | NPU |
| HfWhisperEncoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 27.052 ms | 41 - 48 MB | NPU |
| HfWhisperEncoder | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | 36.746 ms | 3 - 52 MB | NPU |
| HfWhisperEncoder | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | 44.398 ms | 40 - 44 MB | NPU |
| HfWhisperEncoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 18.182 ms | 39 - 46 MB | NPU |
| HfWhisperEncoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 14.536 ms | 0 - 10 MB | NPU |
| HfWhisperEncoder | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 15.857 ms | 0 - 0 MB | NPU |
| HfWhisperEncoder | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 37.311 ms | 0 - 0 MB | NPU |
| HfWhisperEncoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 27.405 ms | 1 - 8 MB | NPU |
| HfWhisperEncoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 120.039 ms | 1 - 10 MB | NPU |
| HfWhisperEncoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | 36.495 ms | 1 - 17 MB | NPU |
| HfWhisperEncoder | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | 43.755 ms | 1 - 10 MB | NPU |
| HfWhisperEncoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | 45.253 ms | 0 - 20 MB | NPU |
| HfWhisperEncoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | 97.486 ms | 1 - 10 MB | NPU |
| HfWhisperEncoder | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | 120.039 ms | 1 - 10 MB | NPU |
| HfWhisperEncoder | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | 74.091 ms | 1 - 6 MB | NPU |
| HfWhisperEncoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 18.248 ms | 1 - 10 MB | NPU |
License
- The license for the original implementation of Whisper-Base can be found here.
References
Community
- Join our AI Hub Slack community to collaborate, post questions and learn more about on-device AI.
- For questions or feedback please reach out to us.