Hugging Face's logo

qualcomm
/
Whisper-Tiny

Automatic Speech Recognition
PyTorch
foundation
android
Model card Files
xet
Community
Whisper-Tiny / README.md
qaihm-bot's picture
qaihm-bot
v0.44.0
3c27f64 verified
preview code
|
raw
history blame contribute delete
12.8 kB
 ---
library_name: pytorch
license: other
tags:
- foundation
- android
pipeline_tag: automatic-speech-recognition
---
![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_tiny/web-assets/model_demo.png)
# Whisper-Tiny: Optimized for Mobile Deployment
## Transformer-based automatic speech recognition (ASR) model for multilingual transcription and translation available on HuggingFace
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 model is an implementation of Whisper-Tiny found [here](https://github.com/huggingface/transformers/tree/v4.42.3/src/transformers/models/whisper).
This repository provides scripts to run Whisper-Tiny on Qualcomm® devices.
More details on model performance across various devices, can be found
[here](https://aihub.qualcomm.com/models/whisper_tiny).
### 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
| Model | Precision | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit | Target Model
|---|---|---|---|---|---|---|---|---|
| HfWhisperEncoder | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_CONTEXT_BINARY | 57.424 ms | 0 - 9 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_CONTEXT_BINARY | 48.98 ms | 1 - 20 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_CONTEXT_BINARY | 18.207 ms | 0 - 3 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | PRECOMPILED_QNN_ONNX | 20.641 ms | 4 - 45 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_CONTEXT_BINARY | 95.138 ms | 1 - 10 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | SA7255P ADP | Qualcomm® SA7255P | QNN_CONTEXT_BINARY | 57.424 ms | 0 - 9 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_CONTEXT_BINARY | 18.151 ms | 1 - 3 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | SA8295P ADP | Qualcomm® SA8295P | QNN_CONTEXT_BINARY | 41.013 ms | 1 - 17 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_CONTEXT_BINARY | 18.302 ms | 1 - 3 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | SA8775P ADP | Qualcomm® SA8775P | QNN_CONTEXT_BINARY | 95.138 ms | 1 - 10 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_CONTEXT_BINARY | 13.476 ms | 0 - 19 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | PRECOMPILED_QNN_ONNX | 15.987 ms | 19 - 38 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_CONTEXT_BINARY | 9.327 ms | 0 - 17 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | PRECOMPILED_QNN_ONNX | 11.383 ms | 20 - 34 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_CONTEXT_BINARY | 7.562 ms | 0 - 11 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | PRECOMPILED_QNN_ONNX | 9.676 ms | 19 - 30 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_CONTEXT_BINARY | 18.661 ms | 0 - 0 MB | NPU | Use Export Script |
| HfWhisperEncoder | float | Snapdragon X Elite CRD | Snapdragon® X Elite | PRECOMPILED_QNN_ONNX | 19.83 ms | 33 - 33 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_CONTEXT_BINARY | 3.637 ms | 7 - 16 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_CONTEXT_BINARY | 2.632 ms | 10 - 27 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_CONTEXT_BINARY | 2.186 ms | 10 - 13 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | PRECOMPILED_QNN_ONNX | 2.44 ms | 0 - 88 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_CONTEXT_BINARY | 2.686 ms | 2 - 12 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | SA7255P ADP | Qualcomm® SA7255P | QNN_CONTEXT_BINARY | 3.637 ms | 7 - 16 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_CONTEXT_BINARY | 2.164 ms | 7 - 9 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | SA8295P ADP | Qualcomm® SA8295P | QNN_CONTEXT_BINARY | 2.967 ms | 10 - 24 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_CONTEXT_BINARY | 2.138 ms | 10 - 13 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | SA8775P ADP | Qualcomm® SA8775P | QNN_CONTEXT_BINARY | 2.686 ms | 2 - 12 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_CONTEXT_BINARY | 1.647 ms | 0 - 20 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | PRECOMPILED_QNN_ONNX | 1.855 ms | 13 - 33 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_CONTEXT_BINARY | 1.326 ms | 0 - 16 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | PRECOMPILED_QNN_ONNX | 1.509 ms | 0 - 14 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_CONTEXT_BINARY | 1.297 ms | 1 - 12 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | PRECOMPILED_QNN_ONNX | 1.432 ms | 1 - 11 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_CONTEXT_BINARY | 1.916 ms | 10 - 10 MB | NPU | Use Export Script |
| HfWhisperDecoder | float | Snapdragon X Elite CRD | Snapdragon® X Elite | PRECOMPILED_QNN_ONNX | 2.012 ms | 84 - 84 MB | NPU | Use Export Script |
## Installation
Install the package via pip:
```bash
# NOTE: 3.10 <= PYTHON_VERSION < 3.14 is supported.
pip install "qai-hub-models[whisper-tiny]"
```
## Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device
Sign-in to [Qualcomm® AI Hub Workbench](https://workbench.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://workbench.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.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.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.export
```
## How does this work?
This [export script](https://aihub.qualcomm.com/models/whisper_tiny/qai_hub_models/models/Whisper-Tiny/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 import Model
# Load the model
torch_model = Model.from_pretrained()
# Device
device = hub.Device("Samsung Galaxy S25")
# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()
pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])
# Compile model on a specific device
compile_job = hub.submit_compile_job(
model=pt_model,
device=device,
input_specs=torch_model.get_input_spec(),
)
# Get target model to run on-device
target_model = 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
profile_job = hub.submit_profile_job(
model=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
input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
model=target_model,
device=device,
inputs=input_data,
)
on_device_output = 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 Workbench. [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's performance across various devices [here](https://aihub.qualcomm.com/models/whisper_tiny).
Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/)
## License
* The license for the original implementation of Whisper-Tiny can be found
[here](https://github.com/huggingface/transformers/blob/v4.42.3/LICENSE).
## References
* [Robust Speech Recognition via Large-Scale Weak Supervision](https://cdn.openai.com/papers/whisper.pdf)
* [Source Model Implementation](https://github.com/huggingface/transformers/tree/v4.42.3/src/transformers/models/whisper)
## 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:ai-hub-support@qti.qualcomm.com).