Hugging Face's logo

qualcomm
/
Whisper-Medium

Automatic Speech Recognition
PyTorch
foundation
android
Model card Files
xet
 Community

Whisper-Medium: Optimized for Qualcomm Devices

HuggingFace Whisper-Medium 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-Medium 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.

Deploying Whisper-Medium on-device

This model is compatible with the Qualcomm Voice AI SDK. Download the SDK from the Qualcomm Package Manager to deploy this model on-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 1.27.1 Download
PRECOMPILED_QNN_ONNX float Snapdragon® X Elite QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 3 Mobile QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 1 Mobile QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ IQ-9075 QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Mobile QAIRT 2.45, ONNX Runtime 1.27.1 Download
PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Gen 5 Mobile QAIRT 2.45, ONNX Runtime 1.27.1 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® Dragonwing™ QCS8550 (Proxy) QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Qualcomm® SA8775P QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ IQ-9075 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 Snapdragon® 8 Elite Mobile QAIRT 2.45 Download
QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Gen 5 Mobile QAIRT 2.45 Download
VOICE_AI float Snapdragon® X2 Elite QAIRT 2.45 Download
VOICE_AI float Snapdragon® X Elite QAIRT 2.45 Download
VOICE_AI float Snapdragon® 8 Gen 3 Mobile QAIRT 2.45 Download
VOICE_AI float Snapdragon® 8 Gen 1 Mobile QAIRT 2.45 Download
VOICE_AI float Qualcomm® Dragonwing™ QCS8550 (Proxy) QAIRT 2.45 Download
VOICE_AI float Qualcomm® SA8775P QAIRT 2.45 Download
VOICE_AI float Qualcomm® Dragonwing™ IQ-9075 QAIRT 2.45 Download
VOICE_AI float Qualcomm® SA7255P QAIRT 2.45 Download
VOICE_AI float Qualcomm® SA8295P QAIRT 2.45 Download
VOICE_AI float Snapdragon® 8 Elite Mobile QAIRT 2.45 Download
VOICE_AI float Snapdragon® 8 Elite Gen 5 Mobile QAIRT 2.45 Download

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

Model Details

Model Type: Model_use_case.speech_recognition

Model Stats:

  • Model checkpoint: openai/whisper-medium
  • Input resolution: 80x3000 (30 seconds audio)
  • Max decoded sequence length: 200 tokens
  • Number of parameters (encoder): 384M
  • Model size (encoder) (float): 1.54GB
  • Number of parameters (decoder): 385M
  • Model size (decoder) (float): 1.54GB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
decoder PRECOMPILED_QNN_ONNX float Snapdragon® X2 Elite 17.935 ms 159 - 159 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® X Elite 27.786 ms 726 - 726 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 3 Mobile 29.42 ms 201 - 213 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 1 Mobile 58.804 ms 199 - 213 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 33.42 ms 0 - 733 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS8450 58.804 ms 199 - 213 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ IQ-9075 37.857 ms 160 - 322 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ IQ-X7181 27.786 ms 726 - 726 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ Q-8750 24.346 ms 50 - 61 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Mobile 24.346 ms 50 - 61 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Gen 5 Mobile 22.456 ms 155 - 167 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® X2 Elite 18.543 ms 160 - 160 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® X Elite 28.141 ms 160 - 160 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 3 Mobile 28.753 ms 158 - 172 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 1 Mobile 58.722 ms 160 - 176 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ QCS8275 51.545 ms 157 - 166 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ QCS8550 (Proxy) 33.949 ms 159 - 162 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® SA8775P 36.329 ms 157 - 169 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® SA8650P 36.329 ms 157 - 169 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® SA8255P 36.329 ms 157 - 169 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® QCS8450 58.722 ms 160 - 176 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ IQ-9075 37.37 ms 160 - 340 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ IQ-X7181 28.141 ms 160 - 160 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ Q-8750 23.488 ms 105 - 119 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® SA7255P 51.545 ms 157 - 166 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® SA8295P 38.394 ms 152 - 163 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Mobile 23.488 ms 105 - 119 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Gen 5 Mobile 20.729 ms 118 - 128 MB NPU
decoder VOICE_AI float Snapdragon® X2 Elite 18.395 ms 160 - 160 MB NPU
decoder VOICE_AI float Snapdragon® X Elite 28.36 ms 160 - 160 MB NPU
decoder VOICE_AI float Snapdragon® 8 Gen 3 Mobile 28.374 ms 135 - 144 MB NPU
decoder VOICE_AI float Qualcomm® Dragonwing™ QCS8275 51.531 ms 141 - 152 MB NPU
decoder VOICE_AI float Qualcomm® Dragonwing™ QCS8550 (Proxy) 33.317 ms 129 - 131 MB NPU
decoder VOICE_AI float Qualcomm® SA8775P 36.595 ms 139 - 151 MB NPU
decoder VOICE_AI float Qualcomm® SA8650P 36.595 ms 139 - 151 MB NPU
decoder VOICE_AI float Qualcomm® SA8255P 36.595 ms 139 - 151 MB NPU
decoder VOICE_AI float Qualcomm® Dragonwing™ IQ-9075 37.257 ms 160 - 340 MB NPU
decoder VOICE_AI float Qualcomm® Dragonwing™ IQ-X7181 28.36 ms 160 - 160 MB NPU
decoder VOICE_AI float Qualcomm® Dragonwing™ Q-8750 23.573 ms 136 - 150 MB NPU
decoder VOICE_AI float Qualcomm® SA7255P 51.531 ms 141 - 152 MB NPU
decoder VOICE_AI float Qualcomm® SA8295P 38.049 ms 112 - 118 MB NPU
decoder VOICE_AI float Snapdragon® 8 Elite Mobile 23.573 ms 136 - 150 MB NPU
decoder VOICE_AI float Snapdragon® 8 Elite Gen 5 Mobile 21.752 ms 141 - 156 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® X2 Elite 152.527 ms 528 - 528 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® X Elite 366.317 ms 698 - 698 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 3 Mobile 274.58 ms 302 - 313 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 1 Mobile 772.763 ms 203 - 216 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 365.769 ms 0 - 770 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS8450 772.763 ms 203 - 216 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ IQ-9075 440.468 ms 161 - 165 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ IQ-X7181 366.317 ms 698 - 698 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ Q-8750 232.322 ms 306 - 317 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Mobile 232.322 ms 306 - 317 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Gen 5 Mobile 152.794 ms 304 - 316 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® X2 Elite 154.371 ms 0 - 0 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® X Elite 366.294 ms 0 - 0 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 3 Mobile 274.085 ms 1 - 14 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 1 Mobile 762.049 ms 0 - 13 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ QCS8275 1339.834 ms 0 - 8 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ QCS8550 (Proxy) 363.019 ms 0 - 2 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® SA8775P 441.561 ms 11 - 22 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® SA8650P 441.561 ms 11 - 22 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® SA8255P 441.561 ms 11 - 22 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® QCS8450 762.049 ms 0 - 13 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ IQ-9075 438.082 ms 0 - 143 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ IQ-X7181 366.294 ms 0 - 0 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ Q-8750 218.424 ms 1 - 13 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® SA7255P 1339.834 ms 0 - 8 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® SA8295P 507.999 ms 3 - 9 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Mobile 218.424 ms 1 - 13 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Gen 5 Mobile 149.77 ms 1 - 10 MB NPU
encoder VOICE_AI float Snapdragon® X2 Elite 154.269 ms 0 - 0 MB NPU
encoder VOICE_AI float Snapdragon® X Elite 364.877 ms 0 - 0 MB NPU
encoder VOICE_AI float Snapdragon® 8 Gen 3 Mobile 274.686 ms 1 - 9 MB NPU
encoder VOICE_AI float Qualcomm® Dragonwing™ QCS8275 1339.099 ms 40 - 50 MB NPU
encoder VOICE_AI float Qualcomm® Dragonwing™ QCS8550 (Proxy) 363.438 ms 0 - 4 MB NPU
encoder VOICE_AI float Qualcomm® SA8775P 441.566 ms 1 - 12 MB NPU
encoder VOICE_AI float Qualcomm® SA8650P 441.566 ms 1 - 12 MB NPU
encoder VOICE_AI float Qualcomm® SA8255P 441.566 ms 1 - 12 MB NPU
encoder VOICE_AI float Qualcomm® Dragonwing™ IQ-9075 439.294 ms 0 - 143 MB NPU
encoder VOICE_AI float Qualcomm® Dragonwing™ IQ-X7181 364.877 ms 0 - 0 MB NPU
encoder VOICE_AI float Qualcomm® Dragonwing™ Q-8750 220.983 ms 0 - 12 MB NPU
encoder VOICE_AI float Qualcomm® SA7255P 1339.099 ms 40 - 50 MB NPU
encoder VOICE_AI float Qualcomm® SA8295P 508.2 ms 1 - 6 MB NPU
encoder VOICE_AI float Snapdragon® 8 Elite Mobile 220.983 ms 0 - 12 MB NPU
encoder VOICE_AI float Snapdragon® 8 Elite Gen 5 Mobile 150.972 ms 1 - 9 MB NPU

License

  • The license for the original implementation of Whisper-Medium 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