Mars Ind - Voice

Mars Ind - Voice Engine is a powerful, flexible text-to-speech tool designed specifically for indie game developers.

A local neural voice studio that runs entirely on your machine — no cloud, no subscription, GPU-accelerated. It pairs two engines in one: Piper TTS drives a wide range of natural voice tones and ages, while XTTS-v2 handles voice cloning and expressive non-word sounds. Reach for the cloning option and both engines work together — that pairing is the heart of the two-engine philosophy.

Key Features

• Multiple AI backends: Piper (lightweight & fast) and XTTS (high-fidelity with voice cloning)

• Age-specific voices (child, adult, elderly) with pitch, tone, and variation control with age gated safety heuristics 

• Emotional & non-verbal sounds (laughs, sighs, "hmm", etc.) with dedicated fine-tuning sliders

• Procedural sound effects (storms, thunder, ambient FX) with looping and randomness controls

• Voice cloning – generate base voices or import samples for consistent characters

• Server mode API support for real-time integration with your AI/LLM

• Export sounds with full metadata for easy Unity integration

Perfect for dynamic NPC dialogue, cutscenes, storytelling, and accessibility features.

Hardware Options

Choose the version that fits your setup:

- NVIDIA (CUDA) for best performance

- AMD (DirectML) for broad compatibility

- CPU fallback for lightweight use

Works great in both standalone desktop mode and as a server for real-time conversations.

Future updates (new voices, features, and improvements) will be free for all owners.

Note: This is Early Access v1. Feedback is highly appreciated and will directly shape future versions.

Created by Mars Ind – Developer Tools for Indie Creators.

AI tools were used as collaborators during implementation, but the vision, structure, and final product are the result of extensive human design and iteration.