--- 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_small/web-assets/model_demo.png) # Whisper-Small: 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-Small found [here](https://github.com/huggingface/transformers/tree/v4.42.3/src/transformers/models/whisper). This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/whisper_small) library to export with custom configurations. More details on model performance across various devices, can be found [here](#performance-summary). Qualcomm AI Hub Models uses [Qualcomm AI Hub Workbench](https://workbench.aihub.qualcomm.com) to compile, profile, and evaluate this model. [Sign up](https://myaccount.qualcomm.com/signup) 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® 8 Elite Gen 5 Mobile | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-precompiled_qnn_onnx-float-qualcomm_snapdragon_8_elite_gen5.zip) | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-precompiled_qnn_onnx-float-qualcomm_snapdragon_x2_elite.zip) | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-precompiled_qnn_onnx-float-qualcomm_snapdragon_x_elite.zip) | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-precompiled_qnn_onnx-float-qualcomm_snapdragon_8gen3.zip) | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-precompiled_qnn_onnx-float-qualcomm_qcs8550_proxy.zip) | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-precompiled_qnn_onnx-float-qualcomm_snapdragon_8_elite_for_galaxy.zip) | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | QAIRT 2.42, ONNX Runtime 1.24.3 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-precompiled_qnn_onnx-float-qualcomm_qcs9075.zip) | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-qnn_context_binary-float-qualcomm_snapdragon_8_elite_gen5.zip) | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-qnn_context_binary-float-qualcomm_snapdragon_x2_elite.zip) | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-qnn_context_binary-float-qualcomm_snapdragon_x_elite.zip) | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-qnn_context_binary-float-qualcomm_snapdragon_8gen3.zip) | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-qnn_context_binary-float-qualcomm_qcs8550_proxy.zip) | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-qnn_context_binary-float-qualcomm_sa8775p.zip) | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-qnn_context_binary-float-qualcomm_snapdragon_8_elite_for_galaxy.zip) | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-qnn_context_binary-float-qualcomm_sa7255p.zip) | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-qnn_context_binary-float-qualcomm_sa8295p.zip) | QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-qnn_context_binary-float-qualcomm_qcs9075.zip) | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | QAIRT 2.45 | [Download](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/whisper_small/releases/v0.51.0/whisper_small-qnn_context_binary-float-qualcomm_qcs8450_proxy.zip) For more device-specific assets and performance metrics, visit **[Whisper-Small on Qualcomm® AI Hub](https://aihub.qualcomm.com/models/whisper_small)**. ### Option 2: Export with Custom Configurations Use the [Qualcomm® AI Hub Models](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/whisper_small) 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-Small on GitHub](https://github.com/qualcomm/ai-hub-models/blob/main/src/qai_hub_models/models/whisper_small) for usage instructions. ## Model Details **Model Type:** Model_use_case.speech_recognition **Model Stats:** - Model checkpoint: openai/whisper-small - Input resolution: 80x3000 (30 seconds audio) - Max decoded sequence length: 200 tokens - Number of parameters (encoder): 102M - Model size (encoder) (float): 391 MB - Number of parameters (decoder): 139M - Model size (decoder) (float): 533 MB ## Performance Summary | Model | Runtime | Precision | Chipset | Inference Time (ms) | Peak Memory Range (MB) | Primary Compute Unit |---|---|---|---|---|---|--- | decoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 7.239 ms | 75 - 85 MB | NPU | decoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 6.06 ms | 286 - 286 MB | NPU | decoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 10.901 ms | 285 - 285 MB | NPU | decoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 9.89 ms | 74 - 85 MB | NPU | decoder | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | 12.214 ms | 61 - 64 MB | NPU | decoder | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | 13.272 ms | 60 - 122 MB | NPU | decoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 8.355 ms | 60 - 67 MB | NPU | decoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 7.347 ms | 60 - 70 MB | NPU | decoder | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 6.976 ms | 60 - 60 MB | NPU | decoder | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 9.93 ms | 60 - 60 MB | NPU | decoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 9.815 ms | 60 - 67 MB | NPU | decoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 18.948 ms | 49 - 56 MB | NPU | decoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | 12.065 ms | 32 - 40 MB | NPU | decoder | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | 13.703 ms | 55 - 64 MB | NPU | decoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | 13.462 ms | 60 - 128 MB | NPU | decoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | 17.306 ms | 60 - 69 MB | NPU | decoder | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | 18.948 ms | 49 - 56 MB | NPU | decoder | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | 14.39 ms | 48 - 53 MB | NPU | decoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 8.354 ms | 0 - 9 MB | NPU | encoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite Gen 5 Mobile | 46.165 ms | 87 - 97 MB | NPU | encoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® X2 Elite | 47.655 ms | 202 - 202 MB | NPU | encoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® X Elite | 115.731 ms | 201 - 201 MB | NPU | encoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Gen 3 Mobile | 81.878 ms | 106 - 113 MB | NPU | encoder | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS8550 (Proxy) | 113.115 ms | 0 - 225 MB | NPU | encoder | PRECOMPILED_QNN_ONNX | float | Qualcomm® QCS9075 | 137.194 ms | 128 - 132 MB | NPU | encoder | PRECOMPILED_QNN_ONNX | float | Snapdragon® 8 Elite For Galaxy Mobile | 59.021 ms | 129 - 136 MB | NPU | encoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite Gen 5 Mobile | 58.968 ms | 1 - 10 MB | NPU | encoder | QNN_CONTEXT_BINARY | float | Snapdragon® X2 Elite | 60.085 ms | 0 - 0 MB | NPU | encoder | QNN_CONTEXT_BINARY | float | Snapdragon® X Elite | 131.993 ms | 0 - 0 MB | NPU | encoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Gen 3 Mobile | 97.234 ms | 3 - 10 MB | NPU | encoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8275 (Proxy) | 449.524 ms | 0 - 8 MB | NPU | encoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8550 (Proxy) | 129.575 ms | 0 - 3 MB | NPU | encoder | QNN_CONTEXT_BINARY | float | Qualcomm® SA8775P | 154.829 ms | 1 - 10 MB | NPU | encoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS9075 | 154.882 ms | 2 - 57 MB | NPU | encoder | QNN_CONTEXT_BINARY | float | Qualcomm® QCS8450 (Proxy) | 272.973 ms | 1 - 9 MB | NPU | encoder | QNN_CONTEXT_BINARY | float | Qualcomm® SA7255P | 449.524 ms | 0 - 8 MB | NPU | encoder | QNN_CONTEXT_BINARY | float | Qualcomm® SA8295P | 200.107 ms | 1 - 6 MB | NPU | encoder | QNN_CONTEXT_BINARY | float | Snapdragon® 8 Elite For Galaxy Mobile | 70.369 ms | 1 - 13 MB | NPU ## License * The license for the original implementation of Whisper-Small 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).