Hugging Face's logo

qualcomm
/
Whisper-Base

Automatic Speech Recognition
PyTorch
foundation
android
Model card Files
xet
 Community
1

Whisper-Base: Optimized for Qualcomm Devices

HuggingFace Whisper-Base 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.

Deploying Whisper-Base 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 1.25.0 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® Dragonwing™ 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® Dragonwing™ IQ-9075 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® Dragonwing™ 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® Dragonwing™ IQ-9075 QAIRT 2.45 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 (encoder): 23.7M
  • Model size (encoder) (float): 90.7 MB
  • Number of parameters (decoder): 48.9M
  • Model size (decoder) (float): 187 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 2.445 ms 20 - 20 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® X Elite 3.715 ms 126 - 126 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 3 Mobile 3.472 ms 16 - 28 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 1 Mobile 5.229 ms 25 - 39 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 4.355 ms 0 - 141 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS8450 5.229 ms 25 - 39 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Gen 5 Mobile 2.684 ms 23 - 33 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ IQ-9075 5.154 ms 20 - 43 MB NPU
decoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Mobile 2.873 ms 2 - 13 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ Q-8750 2.873 ms 2 - 13 MB NPU
decoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ IQ-X7181 3.715 ms 126 - 126 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® X2 Elite 2.846 ms 20 - 20 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® X Elite 3.676 ms 20 - 20 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 3 Mobile 3.366 ms 1 - 9 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 1 Mobile 5.082 ms 19 - 29 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® QCS8275 7.13 ms 20 - 28 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ QCS8550 (Proxy) 4.157 ms 20 - 22 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® SA8775P 5.307 ms 20 - 29 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® SA8650P 5.307 ms 20 - 29 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® SA8255P 5.307 ms 20 - 29 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® QCS8450 5.082 ms 19 - 29 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Gen 5 Mobile 2.657 ms 20 - 29 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ IQ-9075 5.049 ms 20 - 44 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® SA7255P 7.13 ms 20 - 28 MB NPU
decoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Mobile 2.85 ms 1 - 14 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® SA8295P 5.175 ms 18 - 23 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ Q-8750 2.85 ms 1 - 14 MB NPU
decoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ IQ-X7181 3.676 ms 20 - 20 MB NPU
decoder VOICE_AI float Snapdragon® X2 Elite 2.825 ms 20 - 20 MB NPU
decoder VOICE_AI float Snapdragon® X Elite 4.005 ms 20 - 20 MB NPU
decoder VOICE_AI float Snapdragon® 8 Gen 3 Mobile 3.385 ms 1 - 8 MB NPU
decoder VOICE_AI float Snapdragon® 8 Gen 1 Mobile 5.112 ms 19 - 29 MB NPU
decoder VOICE_AI float Qualcomm® QCS8275 7.184 ms 15 - 22 MB NPU
decoder VOICE_AI float Qualcomm® Dragonwing™ QCS8550 (Proxy) 4.239 ms 20 - 22 MB NPU
decoder VOICE_AI float Qualcomm® SA8775P 5.276 ms 20 - 29 MB NPU
decoder VOICE_AI float Qualcomm® SA8650P 5.276 ms 20 - 29 MB NPU
decoder VOICE_AI float Qualcomm® SA8255P 5.276 ms 20 - 29 MB NPU
decoder VOICE_AI float Qualcomm® QCS8450 5.112 ms 19 - 29 MB NPU
decoder VOICE_AI float Snapdragon® 8 Elite Gen 5 Mobile 2.693 ms 20 - 29 MB NPU
decoder VOICE_AI float Qualcomm® Dragonwing™ IQ-9075 5.046 ms 20 - 44 MB NPU
decoder VOICE_AI float Qualcomm® SA7255P 7.184 ms 15 - 22 MB NPU
decoder VOICE_AI float Snapdragon® 8 Elite Mobile 2.846 ms 0 - 9 MB NPU
decoder VOICE_AI float Qualcomm® SA8295P 5.296 ms 18 - 23 MB NPU
decoder VOICE_AI float Qualcomm® Dragonwing™ Q-8750 2.846 ms 0 - 9 MB NPU
decoder VOICE_AI float Qualcomm® Dragonwing™ IQ-X7181 4.005 ms 20 - 20 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® X2 Elite 22.842 ms 67 - 67 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® X Elite 48.796 ms 66 - 66 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 3 Mobile 36.17 ms 39 - 46 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 1 Mobile 95.763 ms 36 - 49 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ QCS8550 (Proxy) 47.735 ms 38 - 39 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS8450 95.763 ms 36 - 49 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Gen 5 Mobile 22.02 ms 39 - 46 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ IQ-9075 70.593 ms 39 - 42 MB NPU
encoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Mobile 25.856 ms 39 - 46 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ Q-8750 25.856 ms 39 - 46 MB NPU
encoder PRECOMPILED_QNN_ONNX float Qualcomm® Dragonwing™ IQ-X7181 48.796 ms 66 - 66 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® X2 Elite 22.699 ms 0 - 0 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® X Elite 49.235 ms 0 - 0 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 3 Mobile 36.012 ms 1 - 8 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 1 Mobile 95.481 ms 0 - 10 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® QCS8275 142.051 ms 1 - 9 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ QCS8550 (Proxy) 47.842 ms 1 - 21 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® SA8775P 55.059 ms 1 - 10 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® SA8650P 55.059 ms 1 - 10 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® SA8255P 55.059 ms 1 - 10 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® QCS8450 95.481 ms 0 - 10 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Gen 5 Mobile 22.028 ms 0 - 14 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ IQ-9075 55.46 ms 1 - 20 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® SA7255P 142.051 ms 1 - 9 MB NPU
encoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Mobile 26.032 ms 1 - 14 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® SA8295P 68.791 ms 0 - 6 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ Q-8750 26.032 ms 1 - 14 MB NPU
encoder QNN_CONTEXT_BINARY float Qualcomm® Dragonwing™ IQ-X7181 49.235 ms 0 - 0 MB NPU
encoder VOICE_AI float Snapdragon® X2 Elite 22.719 ms 0 - 0 MB NPU
encoder VOICE_AI float Snapdragon® X Elite 49.048 ms 0 - 0 MB NPU
encoder VOICE_AI float Snapdragon® 8 Gen 3 Mobile 35.932 ms 1 - 8 MB NPU
encoder VOICE_AI float Snapdragon® 8 Gen 1 Mobile 95.46 ms 0 - 10 MB NPU
encoder VOICE_AI float Qualcomm® QCS8275 142.267 ms 1 - 10 MB NPU
encoder VOICE_AI float Qualcomm® Dragonwing™ QCS8550 (Proxy) 47.694 ms 1 - 17 MB NPU
encoder VOICE_AI float Qualcomm® SA8775P 54.87 ms 1 - 10 MB NPU
encoder VOICE_AI float Qualcomm® SA8650P 54.87 ms 1 - 10 MB NPU
encoder VOICE_AI float Qualcomm® SA8255P 54.87 ms 1 - 10 MB NPU
encoder VOICE_AI float Qualcomm® QCS8450 95.46 ms 0 - 10 MB NPU
encoder VOICE_AI float Snapdragon® 8 Elite Gen 5 Mobile 22.075 ms 1 - 10 MB NPU
encoder VOICE_AI float Qualcomm® Dragonwing™ IQ-9075 55.558 ms 0 - 20 MB NPU
encoder VOICE_AI float Qualcomm® SA7255P 142.267 ms 1 - 10 MB NPU
encoder VOICE_AI float Snapdragon® 8 Elite Mobile 25.686 ms 1 - 10 MB NPU
encoder VOICE_AI float Qualcomm® SA8295P 68.807 ms 1 - 6 MB NPU
encoder VOICE_AI float Qualcomm® Dragonwing™ Q-8750 25.686 ms 1 - 10 MB NPU
encoder VOICE_AI float Qualcomm® Dragonwing™ IQ-X7181 49.048 ms 0 - 0 MB NPU

License

  • The license for the original implementation of Whisper-Base 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