Changing Models into Raspberry Pi 5 Projects

Hi All,

I’m very new to Hailo and Raspberry Pi. We’ve added the Hailo-8 module, which I believe comes with some standard models. However, how do we go about installing additional or custom models? I assume these models are based on available datasets.

Also, is the Hailo-8 strictly limited to video-based AI, or is there any possibility of using it for audio detection as well?

We’re also developing our application using Python, so compatibility with Python-based examples would be important for us.

Hey @user254,

Welcome to the Hailo community!

For pre-trained models, definitely check out our model zoo: GitHub - hailo-ai/hailo_model_zoo: The Hailo Model Zoo includes pre-trained models and a full building and evaluation environment - you’ll find plenty of options that work with both Hailo-8 and Hailo-8L.

If you’re looking to build applications, we’ve got the hailo-apps-infra repository which is Python-based: GitHub - hailo-ai/hailo-apps-infra

And since you mentioned speech recognition, we actually have a working example for that too: Hailo-Application-Code-Examples/runtime/hailo-8/python/speech_recognition at main · hailo-ai/Hailo-Application-Code-Examples · GitHub

Let me know if you need help getting started with any of these!

@omria Thank you for your reply

I believe the Zoo Model is different to the hailo-rpi5-examples?

There are many open-source and royalty-free datasets available — how can these be used with Hailo?
For example, if I only want to detect different types of fruit, I don’t need general object detection for all types of objects.

Currently, we have the Waveshare IMX290-83 camera installed on our Raspberry Pi. I’m not sure what the issue is, but when an object is placed slightly further away (covering fewer pixels on the screen), it doesn’t seem to be detected.

For instance, I would like to detect a person or animal at a distance using the examples provided in the hailo-rpi5-examples repository.

Regarding audio, I’m exploring general audio detection and would like to experiment with identifying sounds such as an animal, car, person, or bike. Is this achievable with Hailo?

Lastly, does Hailo provide direct support for development inquiries, or is the community forum the only available channel?