README.md

---
library_name: pytorch
license: mit
pipeline_tag: automatic-speech-recognition
tags:
- foundation
- android

---

![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_tiny_en/web-assets/model_demo.png)

# Whisper-Tiny-En: Optimized for Mobile Deployment
## Automatic speech recognition (ASR) model for English transcription as well as translation


OpenAI’s Whisper ASR (Automatic Speech Recognition) model is a state-of-the-art system designed for transcribing spoken language into written text. 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 mean decoded length specified below.

This model is an implementation of Whisper-Tiny-En found [here](https://github.com/openai/whisper/tree/main).


This repository provides scripts to run Whisper-Tiny-En on Qualcomm® devices.
More details on model performance across various devices, can be found
[here](https://aihub.qualcomm.com/models/whisper_tiny_en).


### Model Details

- **Model Type:** Speech recognition
- **Model Stats:**
  - Model checkpoint: tiny.en
  - Input resolution: 80x3000 (30 seconds audio)
  - Mean decoded sequence length: 112 tokens
  - Number of parameters (WhisperEncoder): 9.39M
  - Model size (WhisperEncoder): 35.9 MB
  - Number of parameters (WhisperDecoder): 28.2M
  - Model size (WhisperDecoder): 108 MB

| Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
|---|---|---|---|---|---|---|---|---|
| WhisperDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 4.05 ms | 3 - 40 MB | FP16 | NPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperDecoder.tflite) |
| WhisperDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 2.401 ms | 0 - 121 MB | FP16 | NPU | [Whisper-Tiny-En.so](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperDecoder.so) |
| WhisperDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 3.05 ms | 0 - 62 MB | FP16 | NPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperDecoder.tflite) |
| WhisperDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 1.71 ms | 4 - 49 MB | FP16 | NPU | [Whisper-Tiny-En.so](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperDecoder.so) |
| WhisperDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 2.62 ms | 0 - 56 MB | FP16 | NPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperDecoder.tflite) |
| WhisperDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.546 ms | 0 - 42 MB | FP16 | NPU | Use Export Script |
| WhisperDecoder | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 4.107 ms | 3 - 41 MB | FP16 | NPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperDecoder.tflite) |
| WhisperDecoder | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 2.368 ms | 10 - 12 MB | FP16 | NPU | Use Export Script |
| WhisperDecoder | SA7255P ADP | SA7255P | TFLITE | 18.307 ms | 2 - 57 MB | FP16 | NPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperDecoder.tflite) |
| WhisperDecoder | SA7255P ADP | SA7255P | QNN | 15.533 ms | 9 - 19 MB | FP16 | NPU | Use Export Script |
| WhisperDecoder | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 4.007 ms | 3 - 40 MB | FP16 | NPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperDecoder.tflite) |
| WhisperDecoder | SA8255 (Proxy) | SA8255P Proxy | QNN | 2.339 ms | 3 - 4 MB | FP16 | NPU | Use Export Script |
| WhisperDecoder | SA8295P ADP | SA8295P | TFLITE | 5.188 ms | 3 - 55 MB | FP16 | NPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperDecoder.tflite) |
| WhisperDecoder | SA8295P ADP | SA8295P | QNN | 3.586 ms | 1 - 7 MB | FP16 | NPU | Use Export Script |
| WhisperDecoder | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 4.064 ms | 3 - 41 MB | FP16 | NPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperDecoder.tflite) |
| WhisperDecoder | SA8650 (Proxy) | SA8650P Proxy | QNN | 2.422 ms | 10 - 11 MB | FP16 | NPU | Use Export Script |
| WhisperDecoder | SA8775P ADP | SA8775P | TFLITE | 5.443 ms | 0 - 55 MB | FP16 | NPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperDecoder.tflite) |
| WhisperDecoder | SA8775P ADP | SA8775P | QNN | 3.465 ms | 9 - 15 MB | FP16 | NPU | Use Export Script |
| WhisperDecoder | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 4.664 ms | 3 - 62 MB | FP16 | NPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperDecoder.tflite) |
| WhisperDecoder | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 2.892 ms | 4 - 50 MB | FP16 | NPU | Use Export Script |
| WhisperDecoder | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.285 ms | 10 - 10 MB | FP16 | NPU | Use Export Script |
| WhisperEncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 99.892 ms | 20 - 51 MB | FP16 | GPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperEncoder.tflite) |
| WhisperEncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 139.106 ms | 0 - 55 MB | FP16 | NPU | [Whisper-Tiny-En.so](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperEncoder.so) |
| WhisperEncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 81.683 ms | 17 - 47 MB | FP16 | GPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperEncoder.tflite) |
| WhisperEncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 109.268 ms | 0 - 191 MB | FP16 | NPU | [Whisper-Tiny-En.so](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperEncoder.so) |
| WhisperEncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 74.784 ms | 25 - 44 MB | FP16 | GPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperEncoder.tflite) |
| WhisperEncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 98.55 ms | 0 - 195 MB | FP16 | NPU | Use Export Script |
| WhisperEncoder | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 96.819 ms | 13 - 55 MB | FP16 | GPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperEncoder.tflite) |
| WhisperEncoder | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 104.143 ms | 0 - 5 MB | FP16 | NPU | Use Export Script |
| WhisperEncoder | SA7255P ADP | SA7255P | TFLITE | 507.645 ms | 20 - 45 MB | FP16 | GPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperEncoder.tflite) |
| WhisperEncoder | SA7255P ADP | SA7255P | QNN | 464.481 ms | 1 - 10 MB | FP16 | NPU | Use Export Script |
| WhisperEncoder | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 100.562 ms | 18 - 147 MB | FP16 | GPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperEncoder.tflite) |
| WhisperEncoder | SA8255 (Proxy) | SA8255P Proxy | QNN | 106.657 ms | 0 - 5 MB | FP16 | NPU | Use Export Script |
| WhisperEncoder | SA8295P ADP | SA8295P | TFLITE | 103.764 ms | 21 - 42 MB | FP16 | GPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperEncoder.tflite) |
| WhisperEncoder | SA8295P ADP | SA8295P | QNN | 127.685 ms | 4 - 10 MB | FP16 | NPU | Use Export Script |
| WhisperEncoder | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 109.307 ms | 20 - 60 MB | FP16 | GPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperEncoder.tflite) |
| WhisperEncoder | SA8650 (Proxy) | SA8650P Proxy | QNN | 107.644 ms | 0 - 5 MB | FP16 | NPU | Use Export Script |
| WhisperEncoder | SA8775P ADP | SA8775P | TFLITE | 177.953 ms | 20 - 47 MB | FP16 | GPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperEncoder.tflite) |
| WhisperEncoder | SA8775P ADP | SA8775P | QNN | 119.437 ms | 0 - 6 MB | FP16 | NPU | Use Export Script |
| WhisperEncoder | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 185.481 ms | 11 - 48 MB | FP16 | GPU | [Whisper-Tiny-En.tflite](https://huggingface.co/qualcomm/Whisper-Tiny-En/blob/main/WhisperEncoder.tflite) |
| WhisperEncoder | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 165.323 ms | 0 - 196 MB | FP16 | NPU | Use Export Script |
| WhisperEncoder | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 97.277 ms | 0 - 0 MB | FP16 | NPU | Use Export Script |




## Installation

This model can be installed as a Python package via pip.

```bash
pip install "qai-hub-models[whisper_tiny_en]"
```



## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device

Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your
Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`.

With this API token, you can configure your client to run models on the cloud
hosted devices.
```bash
qai-hub configure --api_token API_TOKEN
```
Navigate to [docs](https://app.aihub.qualcomm.com/docs/) for more information.



## Demo off target

The package contains a simple end-to-end demo that downloads pre-trained
weights and runs this model on a sample input.

```bash
python -m qai_hub_models.models.whisper_tiny_en.demo
```

The above demo runs a reference implementation of pre-processing, model
inference, and post processing.

**NOTE**: If you want running in a Jupyter Notebook or Google Colab like
environment, please add the following to your cell (instead of the above).
```
%run -m qai_hub_models.models.whisper_tiny_en.demo
```


### Run model on a cloud-hosted device

In addition to the demo, you can also run the model on a cloud-hosted Qualcomm®
device. This script does the following:
* Performance check on-device on a cloud-hosted device
* Downloads compiled assets that can be deployed on-device for Android.
* Accuracy check between PyTorch and on-device outputs.

```bash
python -m qai_hub_models.models.whisper_tiny_en.export
```
```
Profiling Results
------------------------------------------------------------
WhisperDecoder
Device                          : Samsung Galaxy S23 (13)
Runtime                         : TFLITE                 
Estimated inference time (ms)   : 4.0                    
Estimated peak memory usage (MB): [3, 40]                
Total # Ops                     : 557                    
Compute Unit(s)                 : NPU (557 ops)          

------------------------------------------------------------
WhisperEncoder
Device                          : Samsung Galaxy S23 (13)   
Runtime                         : TFLITE                    
Estimated inference time (ms)   : 99.9                      
Estimated peak memory usage (MB): [20, 51]                  
Total # Ops                     : 271                       
Compute Unit(s)                 : GPU (260 ops) CPU (11 ops)
```


## How does this work?

This [export script](https://aihub.qualcomm.com/models/whisper_tiny_en/qai_hub_models/models/Whisper-Tiny-En/export.py)
leverages [Qualcomm® AI Hub](https://aihub.qualcomm.com/) to optimize, validate, and deploy this model
on-device. Lets go through each step below in detail:

Step 1: **Compile model for on-device deployment**

To compile a PyTorch model for on-device deployment, we first trace the model
in memory using the `jit.trace` and then call the `submit_compile_job` API.

```python
import torch

import qai_hub as hub
from qai_hub_models.models.whisper_tiny_en import Model

# Load the model
model = Model.from_pretrained()
decoder_model = model.decoder
encoder_model = model.encoder

# Device
device = hub.Device("Samsung Galaxy S23")

# Trace model
decoder_input_shape = decoder_model.get_input_spec()
decoder_sample_inputs = decoder_model.sample_inputs()

traced_decoder_model = torch.jit.trace(decoder_model, [torch.tensor(data[0]) for _, data in decoder_sample_inputs.items()])

# Compile model on a specific device
decoder_compile_job = hub.submit_compile_job(
    model=traced_decoder_model ,
    device=device,
    input_specs=decoder_model.get_input_spec(),
)

# Get target model to run on-device
decoder_target_model = decoder_compile_job.get_target_model()
# Trace model
encoder_input_shape = encoder_model.get_input_spec()
encoder_sample_inputs = encoder_model.sample_inputs()

traced_encoder_model = torch.jit.trace(encoder_model, [torch.tensor(data[0]) for _, data in encoder_sample_inputs.items()])

# Compile model on a specific device
encoder_compile_job = hub.submit_compile_job(
    model=traced_encoder_model ,
    device=device,
    input_specs=encoder_model.get_input_spec(),
)

# Get target model to run on-device
encoder_target_model = encoder_compile_job.get_target_model()

```


Step 2: **Performance profiling on cloud-hosted device**

After compiling models from step 1. Models can be profiled model on-device using the
`target_model`. Note that this scripts runs the model on a device automatically
provisioned in the cloud.  Once the job is submitted, you can navigate to a
provided job URL to view a variety of on-device performance metrics.
```python
decoder_profile_job = hub.submit_profile_job(
    model=decoder_target_model,
    device=device,
)
encoder_profile_job = hub.submit_profile_job(
    model=encoder_target_model,
    device=device,
)

```

Step 3: **Verify on-device accuracy**

To verify the accuracy of the model on-device, you can run on-device inference
on sample input data on the same cloud hosted device.
```python
decoder_input_data = decoder_model.sample_inputs()
decoder_inference_job = hub.submit_inference_job(
    model=decoder_target_model,
    device=device,
    inputs=decoder_input_data,
)
decoder_inference_job.download_output_data()
encoder_input_data = encoder_model.sample_inputs()
encoder_inference_job = hub.submit_inference_job(
    model=encoder_target_model,
    device=device,
    inputs=encoder_input_data,
)
encoder_inference_job.download_output_data()

```
With the output of the model, you can compute like PSNR, relative errors or
spot check the output with expected output.

**Note**: This on-device profiling and inference requires access to Qualcomm®
AI Hub. [Sign up for access](https://myaccount.qualcomm.com/signup).




## Deploying compiled model to Android


The models can be deployed using multiple runtimes:
- TensorFlow Lite (`.tflite` export): [This
  tutorial](https://www.tensorflow.org/lite/android/quickstart) provides a
  guide to deploy the .tflite model in an Android application.


- QNN (`.so` export ): This [sample
  app](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-50/sample_app.html)
provides instructions on how to use the `.so` shared library  in an Android application.


## View on Qualcomm® AI Hub
Get more details on Whisper-Tiny-En's performance across various devices [here](https://aihub.qualcomm.com/models/whisper_tiny_en).
Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)


## License
* The license for the original implementation of Whisper-Tiny-En can be found [here](https://github.com/openai/whisper/blob/main/LICENSE).
* The license for the compiled assets for on-device deployment can be found [here](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/Qualcomm+AI+Hub+Proprietary+License.pdf)



## References
* [Robust Speech Recognition via Large-Scale Weak Supervision](https://cdn.openai.com/papers/whisper.pdf)
* [Source Model Implementation](https://github.com/openai/whisper/tree/main)



## Community
* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI.
* For questions or feedback please [reach out to us](mailto:[email protected]).