What text-to-speech and speech-to-text options does Leon support?

Leon supports both cloud and offline voice options, so you can choose based on your privacy needs, budget, and language. According to getleon.ai, Leon's text-to-speech (TTS) integrations include Google Cloud, AWS, IBM Watson, CMU Flite, and Alibaba Cloud, while its speech-to-text (STT) integrations include Google Cloud, IBM Watson, and Coqui STT. Microsoft Azure is listed for both TTS and STT, but as "coming soon" rather than available now.

Text-to-speech (TTS) options

Speech-to-text (STT) options

How to choose between cloud and offline

The site frames this as a privacy trade-off: Leon "lives on your server and you decide if you wish to make use of any third party." That means:

• Pick a cloud provider (Google Cloud, AWS, IBM Watson) if you want higher-accuracy speech recognition or natural-sounding voices and are comfortable sending audio or text to a third party.
• Pick an offline option (CMU Flite for TTS, Coqui STT for STT) if you want voice features to run without external API calls, which keeps data on your own server.
• Wait on Alibaba Cloud or Microsoft Azure if you specifically need those providers — both are listed as "coming soon" for TTS and STT.

What this means in practice

Leon is described as an open-source personal assistant built on Node.js, Python, and AI concepts such as NLP, TTS, and STT. The voice layer is modular, so TTS and STT are separate choices — you can mix a cloud TTS with an offline STT, or the reverse, depending on which direction matters more for your setup.

If you're deciding, start with what you already have: an existing Google Cloud or AWS account makes the cloud path the least setup work, while a fully self-hosted preference points to CMU Flite and Coqui STT.