Hugging Face's logo

qualcomm
/
Whisper-Tiny

Automatic Speech Recognition
PyTorch
foundation
android
Model card Files
xet
Community

Whisper-Tiny: 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-Tiny 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® 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 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 Qualcomm® QCS9075 QAIRT 2.42, ONNX Runtime 1.24.1 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® QCS8275 (Proxy) 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 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 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-Tiny 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-Tiny on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.speech_recognition

Model Stats:

  • Model checkpoint: openai/whisper-tiny
  • Input resolution: 80x3000 (30 seconds audio)
  • Max decoded sequence length: 200 tokens
  • Number of parameters (HfWhisperEncoder): 9.39M
  • Model size (HfWhisperEncoder) (float): 35.9 MB
  • Number of parameters (HfWhisperDecoder): 28.4M
  • Model size (HfWhisperDecoder) (float): 109 MB

Performance Summary

Model Runtime Precision Chipset Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit
HfWhisperDecoder PRECOMPILED_QNN_ONNX float Snapdragon® X Elite 1.957 ms 83 - 83 MB NPU
HfWhisperDecoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 3 Mobile 1.731 ms 13 - 20 MB NPU
HfWhisperDecoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS8550 (Proxy) 2.208 ms 0 - 86 MB NPU
HfWhisperDecoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS9075 2.619 ms 10 - 23 MB NPU
HfWhisperDecoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite For Galaxy Mobile 1.405 ms 0 - 7 MB NPU
HfWhisperDecoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Gen 5 Mobile 1.315 ms 0 - 10 MB NPU
HfWhisperDecoder PRECOMPILED_QNN_ONNX float Snapdragon® X2 Elite 0.924 ms 83 - 83 MB NPU
HfWhisperDecoder QNN_CONTEXT_BINARY float Snapdragon® X Elite 1.896 ms 10 - 10 MB NPU
HfWhisperDecoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 3 Mobile 1.659 ms 10 - 18 MB NPU
HfWhisperDecoder QNN_CONTEXT_BINARY float Qualcomm® QCS8275 (Proxy) 3.903 ms 10 - 17 MB NPU
HfWhisperDecoder QNN_CONTEXT_BINARY float Qualcomm® QCS8550 (Proxy) 2.092 ms 10 - 11 MB NPU
HfWhisperDecoder QNN_CONTEXT_BINARY float Qualcomm® SA8775P 2.733 ms 7 - 16 MB NPU
HfWhisperDecoder QNN_CONTEXT_BINARY float Qualcomm® QCS9075 2.546 ms 10 - 23 MB NPU
HfWhisperDecoder QNN_CONTEXT_BINARY float Qualcomm® QCS8450 (Proxy) 2.632 ms 3 - 12 MB NPU
HfWhisperDecoder QNN_CONTEXT_BINARY float Qualcomm® SA7255P 3.903 ms 10 - 17 MB NPU
HfWhisperDecoder QNN_CONTEXT_BINARY float Qualcomm® SA8295P 2.98 ms 8 - 13 MB NPU
HfWhisperDecoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite For Galaxy Mobile 1.361 ms 0 - 9 MB NPU
HfWhisperDecoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Gen 5 Mobile 1.316 ms 0 - 9 MB NPU
HfWhisperDecoder QNN_CONTEXT_BINARY float Snapdragon® X2 Elite 1.288 ms 10 - 10 MB NPU
HfWhisperEncoder PRECOMPILED_QNN_ONNX float Snapdragon® X Elite 18.059 ms 33 - 33 MB NPU
HfWhisperEncoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Gen 3 Mobile 13.057 ms 20 - 27 MB NPU
HfWhisperEncoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS8550 (Proxy) 17.749 ms 12 - 33 MB NPU
HfWhisperEncoder PRECOMPILED_QNN_ONNX float Qualcomm® QCS9075 21.801 ms 19 - 22 MB NPU
HfWhisperEncoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite For Galaxy Mobile 8.922 ms 20 - 27 MB NPU
HfWhisperEncoder PRECOMPILED_QNN_ONNX float Snapdragon® 8 Elite Gen 5 Mobile 7.503 ms 21 - 31 MB NPU
HfWhisperEncoder PRECOMPILED_QNN_ONNX float Snapdragon® X2 Elite 8.005 ms 31 - 31 MB NPU
HfWhisperEncoder QNN_CONTEXT_BINARY float Snapdragon® X Elite 18.28 ms 0 - 0 MB NPU
HfWhisperEncoder QNN_CONTEXT_BINARY float Snapdragon® 8 Gen 3 Mobile 12.973 ms 0 - 8 MB NPU
HfWhisperEncoder QNN_CONTEXT_BINARY float Qualcomm® QCS8275 (Proxy) 56.987 ms 0 - 9 MB NPU
HfWhisperEncoder QNN_CONTEXT_BINARY float Qualcomm® QCS8550 (Proxy) 17.656 ms 0 - 2 MB NPU
HfWhisperEncoder QNN_CONTEXT_BINARY float Qualcomm® SA8775P 94.404 ms 1 - 9 MB NPU
HfWhisperEncoder QNN_CONTEXT_BINARY float Qualcomm® QCS9075 21.972 ms 2 - 13 MB NPU
HfWhisperEncoder QNN_CONTEXT_BINARY float Qualcomm® QCS8450 (Proxy) 46.518 ms 1 - 10 MB NPU
HfWhisperEncoder QNN_CONTEXT_BINARY float Qualcomm® SA7255P 56.987 ms 0 - 9 MB NPU
HfWhisperEncoder QNN_CONTEXT_BINARY float Qualcomm® SA8295P 46.325 ms 0 - 6 MB NPU
HfWhisperEncoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite For Galaxy Mobile 8.885 ms 0 - 9 MB NPU
HfWhisperEncoder QNN_CONTEXT_BINARY float Snapdragon® 8 Elite Gen 5 Mobile 7.284 ms 0 - 10 MB NPU
HfWhisperEncoder QNN_CONTEXT_BINARY float Snapdragon® X2 Elite 8.249 ms 0 - 0 MB NPU

License

  • The license for the original implementation of Whisper-Tiny 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