--- 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_base/web-assets/model_demo.png) # Whisper-Base: 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-Base found [here](https://github.com/huggingface/transformers/tree/v4.42.3/src/transformers/models/whisper). This repository provides scripts to run Whisper-Base on Qualcomm® devices. More details on model performance across various devices, can be found [here](https://aihub.qualcomm.com/models/whisper_base). ### 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 (HfWhisperEncoder): 23.7M - Model size (HfWhisperEncoder) (float): 90.7 MB - Number of parameters (HfWhisperDecoder): 48.9M - Model size (HfWhisperDecoder) (float): 187 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 | 122.517 ms | 1 - 11 MB | NPU | Use Export Script | | HfWhisperEncoder | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_CONTEXT_BINARY | 98.005 ms | 1 - 22 MB | NPU | Use Export Script | | HfWhisperEncoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_CONTEXT_BINARY | 37.094 ms | 0 - 3 MB | NPU | Use Export Script | | HfWhisperEncoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | PRECOMPILED_QNN_ONNX | 43.552 ms | 8 - 73 MB | NPU | Use Export Script | | HfWhisperEncoder | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_CONTEXT_BINARY | 43.524 ms | 1 - 11 MB | NPU | Use Export Script | | HfWhisperEncoder | float | SA7255P ADP | Qualcomm® SA7255P | QNN_CONTEXT_BINARY | 122.517 ms | 1 - 11 MB | NPU | Use Export Script | | HfWhisperEncoder | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_CONTEXT_BINARY | 37.184 ms | 0 - 3 MB | NPU | Use Export Script | | HfWhisperEncoder | float | SA8295P ADP | Qualcomm® SA8295P | QNN_CONTEXT_BINARY | 75.209 ms | 1 - 17 MB | NPU | Use Export Script | | HfWhisperEncoder | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_CONTEXT_BINARY | 37.08 ms | 1 - 3 MB | NPU | Use Export Script | | HfWhisperEncoder | float | SA8775P ADP | Qualcomm® SA8775P | QNN_CONTEXT_BINARY | 43.524 ms | 1 - 11 MB | NPU | Use Export Script | | HfWhisperEncoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_CONTEXT_BINARY | 26.734 ms | 1 - 19 MB | NPU | Use Export Script | | HfWhisperEncoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | PRECOMPILED_QNN_ONNX | 34.136 ms | 41 - 60 MB | NPU | Use Export Script | | HfWhisperEncoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_CONTEXT_BINARY | 18.158 ms | 0 - 14 MB | NPU | Use Export Script | | HfWhisperEncoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | PRECOMPILED_QNN_ONNX | 23.994 ms | 38 - 53 MB | NPU | Use Export Script | | HfWhisperEncoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_CONTEXT_BINARY | 15.238 ms | 0 - 12 MB | NPU | Use Export Script | | HfWhisperEncoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | PRECOMPILED_QNN_ONNX | 22.515 ms | 38 - 49 MB | NPU | Use Export Script | | HfWhisperEncoder | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_CONTEXT_BINARY | 37.481 ms | 0 - 0 MB | NPU | Use Export Script | | HfWhisperEncoder | float | Snapdragon X Elite CRD | Snapdragon® X Elite | PRECOMPILED_QNN_ONNX | 41.935 ms | 66 - 66 MB | NPU | Use Export Script | | HfWhisperDecoder | float | QCS8275 (Proxy) | Qualcomm® QCS8275 (Proxy) | QNN_CONTEXT_BINARY | 6.271 ms | 17 - 26 MB | NPU | Use Export Script | | HfWhisperDecoder | float | QCS8450 (Proxy) | Qualcomm® QCS8450 (Proxy) | QNN_CONTEXT_BINARY | 5.153 ms | 20 - 41 MB | NPU | Use Export Script | | HfWhisperDecoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | QNN_CONTEXT_BINARY | 3.987 ms | 20 - 22 MB | NPU | Use Export Script | | HfWhisperDecoder | float | QCS8550 (Proxy) | Qualcomm® QCS8550 (Proxy) | PRECOMPILED_QNN_ONNX | 4.53 ms | 0 - 142 MB | NPU | Use Export Script | | HfWhisperDecoder | float | QCS9075 (Proxy) | Qualcomm® QCS9075 (Proxy) | QNN_CONTEXT_BINARY | 4.711 ms | 19 - 29 MB | NPU | Use Export Script | | HfWhisperDecoder | float | SA7255P ADP | Qualcomm® SA7255P | QNN_CONTEXT_BINARY | 6.271 ms | 17 - 26 MB | NPU | Use Export Script | | HfWhisperDecoder | float | SA8255 (Proxy) | Qualcomm® SA8255P (Proxy) | QNN_CONTEXT_BINARY | 3.925 ms | 17 - 19 MB | NPU | Use Export Script | | HfWhisperDecoder | float | SA8295P ADP | Qualcomm® SA8295P | QNN_CONTEXT_BINARY | 5.354 ms | 20 - 34 MB | NPU | Use Export Script | | HfWhisperDecoder | float | SA8650 (Proxy) | Qualcomm® SA8650P (Proxy) | QNN_CONTEXT_BINARY | 3.982 ms | 20 - 22 MB | NPU | Use Export Script | | HfWhisperDecoder | float | SA8775P ADP | Qualcomm® SA8775P | QNN_CONTEXT_BINARY | 4.711 ms | 19 - 29 MB | NPU | Use Export Script | | HfWhisperDecoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | QNN_CONTEXT_BINARY | 3.15 ms | 3 - 22 MB | NPU | Use Export Script | | HfWhisperDecoder | float | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 Mobile | PRECOMPILED_QNN_ONNX | 3.416 ms | 26 - 45 MB | NPU | Use Export Script | | HfWhisperDecoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | QNN_CONTEXT_BINARY | 2.61 ms | 0 - 17 MB | NPU | Use Export Script | | HfWhisperDecoder | float | Samsung Galaxy S25 | Snapdragon® 8 Elite For Galaxy Mobile | PRECOMPILED_QNN_ONNX | 2.819 ms | 0 - 11 MB | NPU | Use Export Script | | HfWhisperDecoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | QNN_CONTEXT_BINARY | 2.46 ms | 18 - 30 MB | NPU | Use Export Script | | HfWhisperDecoder | float | Snapdragon 8 Elite Gen 5 QRD | Snapdragon® 8 Elite Gen5 Mobile | PRECOMPILED_QNN_ONNX | 2.693 ms | 21 - 31 MB | NPU | Use Export Script | | HfWhisperDecoder | float | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN_CONTEXT_BINARY | 3.358 ms | 20 - 20 MB | NPU | Use Export Script | | HfWhisperDecoder | float | Snapdragon X Elite CRD | Snapdragon® X Elite | PRECOMPILED_QNN_ONNX | 3.416 ms | 126 - 126 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-base]" ``` ## 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_base.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_base.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_base.export ``` ## How does this work? This [export script](https://aihub.qualcomm.com/models/whisper_base/qai_hub_models/models/Whisper-Base/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_base 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-Base's performance across various devices [here](https://aihub.qualcomm.com/models/whisper_base). Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/) ## License * The license for the original implementation of Whisper-Base can be found [here](https://github.com/huggingface/transformers/blob/v4.42.3/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/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).