Skip to content

Add blog post "Accelerating Whisper on Arm with PyTorch and Hugging Face Transformers" #1975

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Apr 8, 2025
Merged

Add blog post "Accelerating Whisper on Arm with PyTorch and Hugging Face Transformers" #1975

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
39 changes: 39 additions & 0 deletions _posts/2025-04-08-accelerating-whipser-arm-w-transformers.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,39 @@
---
layout: blog_detail
title: "Accelerating Whisper on Arm with PyTorch and Hugging Face Transformers"
author: Pareena Verma, Arm
---

Automatic speech recognition (ASR) has revolutionized how we interact with technology, clearing the way for applications like real-time audio transcription, voice assistants, and accessibility tools. OpenAI Whisper is a powerful model for ASR, capable of multilingual speech recognition and translation.

A new Arm Learning Path is now available that explains how to accelerate Whisper on Arm-based cloud instances using PyTorch and Hugging Face transformers.

**Why Run Whisper on Arm?**

Arm processors are popular in cloud infrastructure for their efficiency, performance, and cost-effectiveness. With major cloud providers such as AWS, Azure, and Google Cloud offering Arm-based instances, running machine learning workloads on this architecture is becoming increasingly attractive.

**What You’ll Learn**

The [Arm Learning Path](https://learn.arm.com/learning-paths/servers-and-cloud-computing/whisper/) provides a structured approach to setting up and accelerating Whisper on Arm-based cloud instances. Here’s what you cover:

**1. Set Up Your Environment**

Before running Whisper, you must set up your development environment. The learning path walks you through setting up an Arm-based cloud instance and installing all dependencies, such as PyTorch, Transformers, and ffmpeg.

**2. Run Whisper with PyTorch and Hugging Face Transformers**

Once the environment is ready, you will use the Hugging Face transformer library with PyTorch to load and execute Whisper for speech-to-text conversion. The tutorial provides a step-by-step approach for processing audio files and generating audio transcripts.

**3. Measure and Evaluate Performance**

To ensure efficient execution, you learn how to measure transcription speeds and compare different optimization techniques. The guide provides insights into interpreting performance metrics and making informed decisions on your deployment.

**Try it Yourself**

Upon completion of this tutorial, you know how to:

* Deploy Whisper on an Arm-based cloud instance.
* Implement performance optimizations for efficient execution.
* Evaluate transcription speeds and optimize further based on results.

**Try the live demo today** and see audio transcription in action on Arm: [Whisper on Arm Demo](https://learn.arm.com/learning-paths/servers-and-cloud-computing/whisper/_demo/).