Run Conformer speech recognition on Alif Ensemble E8 with ExecuTorch
Introduction
Set up Conformer ASR and MLEK resources
Build and run ASR on a Corstone-320 FVP
Prepare the Alif E8 ASR application and model
Build and flash the ASR firmware
Verify live speech recognition and NPU execution
Next Steps
Run Conformer speech recognition on Alif Ensemble E8 with ExecuTorch
Who is this for?
This Learning Path is for embedded ML developers who want to capture and transcribe live speech on an Alif Ensemble E8 using ExecuTorch and the Ethos-U85 NPU.
What will you learn?
After completing this Learning Path, you'll be able to perform the following tasks:
- Prepare a quantized Conformer model with Arm and Alif MLEK for ExecuTorch deployment.
- Build and run ASR on the Corstone-320 Fixed Virtual Platform (FVP) using recorded audio.
- Build and flash the E8 application and its external-flash model.
- Validate live transcription, Ethos-U85 execution, and standalone restart on the E8.
Prerequisites
Before starting, you'll need the following prerequisites:
- Experience with C or C++, CMake, and embedded baremetal development concepts
- Familiarity with PyTorch, ExecuTorch, and quantized model deployment
- A Linux development machine, or an Apple silicon Mac running macOS 15 or later with Homebrew and Docker Desktop installed
- An Alif Ensemble E8 AI/ML AppKit (AK-E8-AIML), or an E8 DevKit with a compatible attached display, and a USB-C data cable
Summary
This summary was drafted with an approved AI-assisted workflow and reviewed by Arm contributors before publication. Human technical review remains part of the process so the final page reflects engineering rigor, accuracy, and Arm editorial standards.
You’ll prepare a quantized Conformer speech recognition model for ExecuTorch, validate it on a Corstone-320 FVP, and deploy it to an Alif Ensemble E8. First, you’ll generate model resources and run the FVP application with recorded audio. Then, you’ll build the E8 firmware, program the model in external flash, and capture speech through the onboard microphones. Finally, you’ll check the displayed transcription, Ethos-U85 activity, and standalone restart.