Realtime synthesis
Stream generated speech as text is converted into audio, suitable for adaptive dialogue or live narration.
Vocalis
Generate natural-sounding speech from text locally with no internet dependency. Output directly to an AudioSource, generate an AudioClip or export generated content through SonicLoad for downstream workflows.

SOURCE PREVIEW
Quick start
What it does
Stream generated speech as text is converted into audio, suitable for adaptive dialogue or live narration.
Produce ready-to-use Unity audio assets or runtime clip output for dynamic scenes and gameplay events.
Adjust pitch, speed and voice styles to fit game tone, character personality or accessibility workflows.
Use multiple multilingual voices from a local library designed for modular games and localization.
Layer it into SonicLoad export flows or internal asset pipelines for recording and serialized playback.
Trigger speech generation from gameplay events, story flow or editor scripts using C# callbacks and Inspector bindings.
Planned API
using Denisitree.Vocalis;
using UnityEngine;
public class DialogueVoice : MonoBehaviour
{
private SpeechSynthesizer synthesizer;
private async void Start()
{
synthesizer = new SpeechSynthesizer();
synthesizer.Voice = "en-US-voice-01";
synthesizer.Speed = 1.0f;
synthesizer.Pitch = 0.0f;
var clip = await synthesizer.SynthesizeAsync("Welcome to the outpost, traveler.");
var source = GetComponent<AudioSource>();
source.clip = clip;
source.Play();
}
}
Planned platforms & requirements
| Category | Planned support |
|---|---|
| Platforms | Windows, macOS, Linux, Android, iOS |
| Unity | Unity 2021.3 LTS+ |
| Output targets | AudioSource, AudioClip, export via SonicLoad |
| Voice options | Large multilingual voice library with style tuning |
| Generation | Async synthesis and streaming-ready architecture |
Works well with
Roadmap
Screenshots