Build offline voice assisstant AI

Job ID: 31867798

Budget: $250 – $750 AUD

I am hoping to develop a solution for AI project.

It is building an Offline Voice assistant (like Google, Siri, Alexa) that is to run on a Raspberry Pi or similar Micro Computer architecture (IoT) which is ARM-based environment.

For the speech to text model, I am want DeepSpeech implementation, and for the text to speech, something from Mozilla TTS. If there are any other AI implementation I am happy to listen to suggestions.

The AI models can be trained on a desktop with GPUs; however, the models themselves need to work on ARM-based environment so the solution must be compatible to run on it! I would recommend TensorFlow lite to be used for the system as it is designed for embedded and microprocessors.