Dear Hailo Support Team,
I am currently studying AI and AI agents, and I am working on building a GenAI voice assistant using Hailo-10H for learning and research purposes.
My development environment:
- Raspberry Pi 5
- Ubuntu 24.04 LTS
- Hailo-10H (PCIe)
- HailoRT 5.3.0
- hailort-pcie-driver 5.3.0
- hailo_gen_ai_model_zoo 5.1.1
I am looking for the package:
hailo-genai-application_x.x.x_arm64.deb
which includes the following applications:
- hailo_whisper_app
- hailo_llm_app
- hailo_genai_app
- hailo_audio_app
However, I could not find this package on the Hailo Developer Zone download page.
Could you please advise how I can obtain this package, or what steps are required to access it?
This is for learning and research purposes.
Thank you very much for your support.
The example applications are available in our GitHub repository:
Hello,
I am working with Raspberry Pi 5 (Ubuntu 24.04 / Python 3.12) + Hailo-10H.
The installation of hailo-apps using install.sh completed successfully (Step 7/7).
However, the following GenAI directories are missing:
hailo-apps/hailo_apps/genai/
├── whisper_service
├── llm_service
├── tts_service
└── assistant_app
These directories do not exist in the current hailo-apps repository,
and install.sh does not generate them.
Question
Where can I obtain the GenAI demo applications
(whisper_service / llm_service / tts_service / assistant_app)?
-
Were they moved to another repository?
-
Are they available in the Developer Zone?
-
Or are they no longer publicly distributed?
I would like to build a fully local voice assistant (Whisper → LLM → TTS),
so I need access to the GenAI demo applications.
Thank you.
The genai apps are here:
Thank you for the information.
I have checked the hailo-apps/python/gen_ai_apps repository and I’m currently working on launching the Voice Assistant application (Whisper → LLM → TTS) on Hailo-10H.
If there are any recommended settings or best practices for integrating the voice_assistant implementation into a custom GenAI application, I would appreciate your guidance.
Thanks again for your support.
Voice-enabled GenAI applications are still an emerging area, so there is not yet a broadly established set of recommended settings or best practices.
At this stage, we are providing an example implementation as a starting point and invite you to experiment with it, adapt it to your own use case, and explore creative approaches. As adoption grows and we learn from real-world implementations, we will continue to refine our guidance and share additional recommendations over time.