Hugging Face's logo

qualcomm
/
Whisper-Small

Automatic Speech Recognition
PyTorch
foundation
android
Model card Files
xet
 Community

Whisper-Small: 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-Small 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® X2 Elite QAIRT 2.45, ONNX Runtime 1.25.0 Download
PRECOMPILED_QNN_ONNX float Snapdragon® X Elite QAIRT 2.45, ONNX Runtime 1.25.0 Download
PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 3 Mobile QAIRT 2.45, ONNX Runtime 1.25.0 Download
PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 1 Mobile QAIRT 2.45, ONNX Runtime 1.25.0 Download
PRECOMPILED_QNN_ONNX float Qualcomm® QCS8550 (Proxy) QAIRT 2.45, ONNX Runtime 1.25.0 Download
PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Mobile QAIRT 2.45, ONNX Runtime 1.25.0 Download
PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Gen 5 Mobile QAIRT 2.45, ONNX Runtime 1.25.0 Download
PRECOMPILED_QNN_ONNX float Qualcomm® QCS9075 QAIRT 2.45, ONNX Runtime 1.25.0 Download
QNN_CONTEXT_BINARY float Snapdragon® X2 Elite QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Snapdragon® X Elite QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 3 Mobile QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 1 Mobile QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Qualcomm® QCS8550 (Proxy) QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Qualcomm® SA8775P QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Mobile QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Gen 5 Mobile QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Qualcomm® SA7255P QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Qualcomm® SA8295P QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Qualcomm® QCS9075 QAIRT 2.45 Download

For more device-specific assets and performance metrics, visit Whisper-Small 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-Small on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.speech_recognition

Model Stats:

  • Model checkpoint: openai/whisper-small
  • Input resolution: 80x3000 (30 seconds audio)
  • Max decoded sequence length: 200 tokens
  • Number of parameters (encoder): 102M
  • Model size (encoder) (float): 391 MB
  • Number of parameters (decoder): 139M
  • Model size (decoder) (float): 533 MB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
decoder PRECOMPILED_QNN_ONNX float Snapdragon® X2 Elite 6.645 ms 60 - 60 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® X Elite 10.07 ms 287 - 287 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 3 Mobile 9.84 ms 75 - 87 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 1 Mobile 18.347 ms 75 - 89 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS8550 (Proxy) 12.343 ms 62 - 64 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS8450 18.347 ms 75 - 89 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Gen 5 Mobile 7.439 ms 43 - 51 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS9075 13.697 ms 60 - 122 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Mobile 8.428 ms 58 - 70 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS8750 8.428 ms 58 - 70 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS7181 10.07 ms 287 - 287 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® X2 Elite 7.088 ms 60 - 60 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® X Elite 10.172 ms 60 - 60 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 3 Mobile 9.695 ms 36 - 44 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 1 Mobile 17.665 ms 60 - 75 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® QCS8275 19.253 ms 48 - 56 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® QCS8550 (Proxy) 12.074 ms 58 - 60 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® QCS8450 17.665 ms 60 - 75 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Gen 5 Mobile 7.365 ms 1 - 10 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® SA7255P 19.253 ms 48 - 56 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® QCS9075 13.422 ms 60 - 128 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Mobile 8.371 ms 7 - 20 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® SA8295P 14.617 ms 58 - 63 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® QCS8750 8.371 ms 7 - 20 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® QCS7181 10.172 ms 60 - 60 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® X2 Elite 59.819 ms 132 - 132 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® X Elite 133.381 ms 255 - 255 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 3 Mobile 97.509 ms 107 - 113 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 1 Mobile 273.643 ms 125 - 140 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS8550 (Proxy) 131.724 ms 0 - 258 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS8450 273.643 ms 125 - 140 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Gen 5 Mobile 58.601 ms 127 - 138 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS9075 155.305 ms 127 - 130 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Mobile 70.427 ms 111 - 123 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS8750 70.427 ms 111 - 123 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS7181 133.381 ms 255 - 255 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® X2 Elite 60.168 ms 0 - 0 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® X Elite 132.887 ms 0 - 0 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 3 Mobile 97.198 ms 17 - 24 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 1 Mobile 270.612 ms 0 - 13 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® QCS8275 449.261 ms 0 - 8 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® QCS8550 (Proxy) 130.318 ms 0 - 3 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® QCS8450 270.612 ms 0 - 13 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Gen 5 Mobile 58.128 ms 1 - 9 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® SA7255P 449.261 ms 0 - 8 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® QCS9075 154.733 ms 0 - 55 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Mobile 70.132 ms 1 - 10 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® SA8295P 200.172 ms 1 - 6 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® QCS8750 70.132 ms 1 - 10 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® QCS7181 132.887 ms 0 - 0 MB NPU

License

  • The license for the original implementation of Whisper-Small can be found here.

References

Community

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
Automatic Speech Recognition
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support