Distil-Whisper

qualcomm/Distil-Whisper

Distil-Whisper is a automatic speech recognition model from qualcomm. Use it when you need speech turned into text. It is set up for pytorch. The card lists the license as mit.

Downloads · 30 days

Access

Public

Updated Sep 23, 2026

Repo size

—

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.md11.6 KB · 81%

At a glance

Task: Automatic Speech Recognition
Library: pytorch
License: mit
Access: Public
Created: Apr 9, 2026
Updated: Sep 23, 2026
SHA: 07b95297

Model card

From the Hugging Face model README

Distil-Whisper: Optimized for Qualcomm Devices

Distil-Whisper Small English is a distilled version of Whisper Small, optimized for fast and efficient automatic speech recognition.

This is based on the implementation of Distil-Whisper 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
ONNX	float	Universal	QAIRT 2.50, ONNX Runtime 1.27.1	Download
QNN_DLC	float	Universal	QAIRT 2.50	Download
TFLITE	float	Universal		Download

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

Model Details

Model Type: Model_use_case.speech_recognition

Model Stats:

Input resolution: 80x3000 (30 seconds audio)
Max decoded sequence length: 200 tokens
Model checkpoint: distil-whisper/distil-small.en
Model size (decoder) (float): 450MB
Model size (encoder) (float): 332 MB
Number of parameters (decoder): 211M
Number of parameters (encoder): 166M

Performance Summary

Model	Runtime	Precision	Chipset	Inference Time (ms)	Peak Memory Range (MB)	Primary Compute Unit
decoder	ONNX	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	5.603 ms	7 - 350 MB	NPU
decoder	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	7.297 ms	15 - 369 MB	NPU
decoder	ONNX	float	Snapdragon® X2 Elite	5.043 ms	40 - 40 MB	NPU
decoder	ONNX	float	Snapdragon® X Elite	11.315 ms	179 - 179 MB	NPU
decoder	ONNX	float	Snapdragon® 8 Gen 3 Mobile	8.88 ms	0 - 472 MB	NPU
decoder	ONNX	float	Snapdragon® 8 Gen 1 Mobile	13.895 ms	50 - 389 MB	NPU
decoder	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	13.01 ms	40 - 83 MB	NPU
decoder	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	11.765 ms	0 - 182 MB	NPU
decoder	ONNX	float	Qualcomm® QCS8450	13.895 ms	50 - 389 MB	NPU
decoder	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	12.832 ms	40 - 83 MB	NPU
decoder	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	11.315 ms	179 - 179 MB	NPU
decoder	ONNX	float	Qualcomm® Dragonwing™ Q-8750	7.297 ms	15 - 369 MB	NPU
decoder	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	5.735 ms	4 - 333 MB	NPU
decoder	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	7.399 ms	0 - 340 MB	NPU
decoder	QNN_DLC	float	Snapdragon® X2 Elite	5.526 ms	40 - 40 MB	NPU
decoder	QNN_DLC	float	Snapdragon® X Elite	11.315 ms	40 - 40 MB	NPU
decoder	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	8.759 ms	0 - 428 MB	NPU
decoder	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	13.625 ms	31 - 320 MB	NPU
decoder	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	12.766 ms	40 - 87 MB	NPU
decoder	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	11.56 ms	40 - 43 MB	NPU
decoder	QNN_DLC	float	Qualcomm® SA8775P	13.175 ms	19 - 346 MB	NPU
decoder	QNN_DLC	float	Qualcomm® SA8650P	13.175 ms	19 - 346 MB	NPU
decoder	QNN_DLC	float	Qualcomm® SA8255P	13.175 ms	19 - 346 MB	NPU
decoder	QNN_DLC	float	Qualcomm® QCS8450	13.625 ms	31 - 320 MB	NPU
decoder	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	12.632 ms	40 - 87 MB	NPU
decoder	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	11.315 ms	40 - 40 MB	NPU
decoder	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	7.399 ms	0 - 340 MB	NPU
decoder	QNN_DLC	float	Qualcomm® SA7255P	19.128 ms	30 - 356 MB	NPU
decoder	QNN_DLC	float	Qualcomm® SA8295P	14.092 ms	5 - 200 MB	NPU
decoder	TFLITE	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	5.838 ms	4 - 384 MB	NPU
decoder	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	7.207 ms	4 - 371 MB	NPU
decoder	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	8.605 ms	5 - 554 MB	NPU
decoder	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	13.966 ms	5 - 454 MB	NPU
decoder	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	12.92 ms	0 - 266 MB	NPU
decoder	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	11.643 ms	5 - 8 MB	NPU
decoder	TFLITE	float	Qualcomm® SA8775P	13.055 ms	5 - 358 MB	NPU
decoder	TFLITE	float	Qualcomm® SA8650P	13.055 ms	5 - 358 MB	NPU
decoder	TFLITE	float	Qualcomm® SA8255P	13.055 ms	5 - 358 MB	NPU
decoder	TFLITE	float	Qualcomm® QCS8450	13.966 ms	5 - 454 MB	NPU
decoder	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	12.822 ms	0 - 265 MB	NPU
decoder	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	7.207 ms	4 - 371 MB	NPU
decoder	TFLITE	float	Qualcomm® SA7255P	19.228 ms	5 - 357 MB	NPU
decoder	TFLITE	float	Qualcomm® SA8295P	13.909 ms	5 - 261 MB	NPU
encoder	ONNX	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	54.074 ms	79 - 648 MB	NPU
encoder	ONNX	float	Snapdragon® 8 Elite For Galaxy Mobile	65.838 ms	78 - 626 MB	NPU
encoder	ONNX	float	Snapdragon® X2 Elite	55.095 ms	133 - 133 MB	NPU
encoder	ONNX	float	Snapdragon® X Elite	118.942 ms	236 - 236 MB	NPU
encoder	ONNX	float	Snapdragon® 8 Gen 3 Mobile	88.727 ms	0 - 922 MB	NPU
encoder	ONNX	float	Snapdragon® 8 Gen 1 Mobile	177.076 ms	76 - 965 MB	NPU
encoder	ONNX	float	Qualcomm® Dragonwing™ IQ-8275	147.426 ms	74 - 79 MB	NPU
encoder	ONNX	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	118.757 ms	0 - 1197 MB	NPU
encoder	ONNX	float	Qualcomm® QCS8450	177.076 ms	76 - 965 MB	NPU
encoder	ONNX	float	Qualcomm® Dragonwing™ IQ-9075	141.055 ms	79 - 83 MB	NPU
encoder	ONNX	float	Qualcomm® Dragonwing™ IQ-X7181	118.942 ms	236 - 236 MB	NPU
encoder	ONNX	float	Qualcomm® Dragonwing™ Q-8750	65.838 ms	78 - 626 MB	NPU
encoder	QNN_DLC	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	54.317 ms	0 - 547 MB	NPU
encoder	QNN_DLC	float	Snapdragon® 8 Elite For Galaxy Mobile	65.358 ms	0 - 526 MB	NPU
encoder	QNN_DLC	float	Snapdragon® X2 Elite	56.123 ms	1 - 1 MB	NPU
encoder	QNN_DLC	float	Snapdragon® X Elite	119.253 ms	1 - 1 MB	NPU
encoder	QNN_DLC	float	Snapdragon® 8 Gen 3 Mobile	87.382 ms	0 - 802 MB	NPU
encoder	QNN_DLC	float	Snapdragon® 8 Gen 1 Mobile	175.959 ms	0 - 766 MB	NPU
encoder	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-8275	145.895 ms	1 - 40 MB	NPU
encoder	QNN_DLC	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	117.014 ms	1 - 3 MB	NPU
encoder	QNN_DLC	float	Qualcomm® SA8775P	139.031 ms	1 - 523 MB	NPU
encoder	QNN_DLC	float	Qualcomm® SA8650P	139.031 ms	1 - 523 MB	NPU
encoder	QNN_DLC	float	Qualcomm® SA8255P	139.031 ms	1 - 523 MB	NPU
encoder	QNN_DLC	float	Qualcomm® QCS8450	175.959 ms	0 - 766 MB	NPU
encoder	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-9075	139.849 ms	1 - 39 MB	NPU
encoder	QNN_DLC	float	Qualcomm® Dragonwing™ IQ-X7181	119.253 ms	1 - 1 MB	NPU
encoder	QNN_DLC	float	Qualcomm® Dragonwing™ Q-8750	65.358 ms	0 - 526 MB	NPU
encoder	QNN_DLC	float	Qualcomm® SA7255P	390.313 ms	1 - 524 MB	NPU
encoder	QNN_DLC	float	Qualcomm® SA8295P	174.471 ms	1 - 501 MB	NPU
encoder	TFLITE	float	Snapdragon® 8 Elite Gen 5 For Galaxy Mobile	363.7 ms	42 - 82 MB	GPU
encoder	TFLITE	float	Snapdragon® 8 Elite For Galaxy Mobile	414.533 ms	40 - 78 MB	GPU
encoder	TFLITE	float	Snapdragon® 8 Gen 3 Mobile	477.419 ms	39 - 182 MB	GPU
encoder	TFLITE	float	Snapdragon® 8 Gen 1 Mobile	846.172 ms	40 - 194 MB	GPU
encoder	TFLITE	float	Qualcomm® Dragonwing™ IQ-8275	2994.928 ms	0 - 39 MB	GPU
encoder	TFLITE	float	Qualcomm® Dragonwing™ QCS8550 (Proxy)	652.165 ms	0 - 318 MB	GPU
encoder	TFLITE	float	Qualcomm® SA8775P	1325.637 ms	38 - 83 MB	GPU
encoder	TFLITE	float	Qualcomm® SA8650P	1325.637 ms	38 - 83 MB	GPU
encoder	TFLITE	float	Qualcomm® SA8255P	1325.637 ms	38 - 83 MB	GPU
encoder	TFLITE	float	Qualcomm® QCS8450	846.172 ms	40 - 194 MB	GPU
encoder	TFLITE	float	Qualcomm® Dragonwing™ IQ-9075	1326.282 ms	0 - 39 MB	GPU
encoder	TFLITE	float	Qualcomm® Dragonwing™ Q-8750	414.533 ms	40 - 78 MB	GPU
encoder	TFLITE	float	Qualcomm® SA7255P	3109.457 ms	18 - 63 MB	GPU
encoder	TFLITE	float	Qualcomm® SA8295P	685.442 ms	38 - 82 MB	GPU

License

The license for the original implementation of Distil-Whisper 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.

Task: Automatic Speech Recognition
Library: pytorch
License: mit
Created: Apr 9, 2026
Updated: Sep 23, 2026