Whisper Workshop 2025, AKL, NZ
September 5, 9AM
We look forward to seeing you there.
Whisper Workshop Practice
Day(s)
:
Hour(s)
:
Minute(s)
:
Second(s)
About the Event
A Deep Dive into the Whisper
Whisper is one of the most advanced speech processing models, enabling state-of-the-art transcription and translation capabilities. This workshop will introduce the fundamentals of speech processing, explore Whisper and Transformer architectures, and provide a hands-on interactive notebook session to help you customize Whisper for your own projects.
đź“Ť Venue: Room 405-328, 20 Symonds Street, Auckland CBD
đź’˛ Cost: Free (refreshments and lunch provided)
🎟 Tickets: Limited to 30 participants – register early
Learning Objectives
By participating in this workshop, you’ll:Â
-
- Understand core concepts in speech processing.
- Explore Whisper’s architecture and why it leads the field.
- Practice using Whisper’s built-in functions and tuning its settings.
- Apply fine-tuning for specialized vocabularies, accents, and audio types.
Workshop details
 Prerequisites: Basic Programming knowledge
Suggested materials to satisfy prerequisites: Python Beginner’s Guide.
Technologies: PyTorch, Jupyter Notebook
Hardware Requirements: Please bring your own laptop
Featured Sessions
Speech Processing
Associate Professor Waleed will open the workshop by presenting the foundations of speech processing to secure a solid base for the entire event.
Fri, Sep 5, 2025
9:20 AM to 10:00 AM
Foundation of Whisper
This session unpacks Whisper’s core architecture and training strategy, explaining the mechanisms that make it a state-of-the-art speech model. By understanding these foundations, you will be fully prepared for the practical exercises that follow.
Fri, Sep 5, 2025
10:00 AM to 12:00 AM
Lunch Time
Take a break, grab a plate, and enjoy lunch with us!
Fri, Sep 5, 2025
12:00 AM to 1:00 PM
Hands-On Whisper Workflows
After the break we will dive into hands-on labs using Whisper’s built-in functions. You will experiment with transcription, translation, and automatic timestamping—capabilities that can facilitate your workflow such as turning interview recordings into text and much more.
Fri, Sep 5, 2025
1:00 PM to 2:00 PM
Whisper Customization
In the last session, we will provide a step-by-step guide on how to fine-tune Whisper with your own custom dataset. You will leave with the skills to build a customized model tailored to your project’s unique needs.
Fri, Sep 5, 2025
2:00 PM to 4:00 PM
Registration
The DeepNet Discovery Network cordially invite you to our upcoming workshop, exclusively for academics and students in the Department of Electrical, Computer and Software Engineering. Please register using your university email address to secure your place.
Keynote speakers
Satwinder Singh
Dr
Waleed Abdulla
Associate Professor
Zihan Zhong
PhD candidateÂ
Ben Wang
PhD Candidate