Skip to content

TTS Models

Tryll synthesizes speech through the SherpaOnnx engine. This page is the authoritative list of which Sherpa-ONNX TTS families the Tryll server can load, which it cannot, and why — so you can tell at a glance whether a voice you found upstream will work here.

Short version: Tryll ships an eSpeak-free Sherpa-ONNX build so the server can be redistributed inside a proprietary game. That choice makes two families available — Supertonic 3 and Pocket TTS — and puts the rest out of reach.

For the catalog file format these models are declared in, see Model Management. For the node parameters that select a voice at runtime, see TTS and Voice Output.

Shipped voices

Both entries below are in the default models.json, are downloaded on demand, and are selected with the tts_model_name parameter on GenerateAndSpeak or Speak.

Catalog name Family Voices Languages Voice cloning Weights license
Supertonic 3 (int8) supertonic 10 built-in styles (speaker_id 0–9) 31, via tts_lang No OpenRAIL-M (attribution + use restrictions)
Pocket TTS (int8) pocket Single-speaker (speaker_id must stay 0) English Yes — zero-shot from a reference WAV (tts_voice) CC-BY-4.0 (attribution)

Pick Supertonic 3 when you want multiple distinct voices, non-English speech, or the lowest latency. Pick Pocket TTS when a character needs a specific voice you can supply as an audio sample — see Clone a Voice.

Attribution is required for both

Neither model's weights are public domain. Supertonic 3 is OpenRAIL-M (attribution plus use-based restrictions — no impersonation or harmful use); Pocket TTS is CC-BY-4.0 (attribution). If you ship either voice in a game, include the model attribution in your credits or third-party notices. The engine code itself (Sherpa-ONNX, ONNX Runtime) is permissively licensed and imposes no such obligation.

Family support matrix

Sherpa-ONNX v1.13.2 exposes seven TTS families. Tryll accepts two of them.

Family Upstream examples Status in Tryll Why
Supertonic Supertonic 3 ✅ Supported Own tokenizer (unicode_indexer.bin); no eSpeak
Pocket Pocket TTS (Kyutai) ✅ Supported SentencePiece vocabulary; no eSpeak
Kokoro Kokoro v0.19, v1.0 ❌ Unavailable Requires eSpeak NG — see below
Kitten KittenTTS nano/micro/mini ❌ Unavailable Requires eSpeak NG — see below
VITS (incl. Piper) Piper voices, MMS, LJSpeech ❌ Unavailable Piper-style bundles require eSpeak NG. Lexicon-based VITS bundles do not, but the family is not implemented in Tryll.
Matcha Matcha-Icefall (en, zh) ❌ Unavailable English bundles require eSpeak NG. The Chinese lexicon path does not, but the family is not implemented in Tryll.
ZipVoice ZipVoice, ZipVoice-Distill ❌ Unavailable No eSpeak requirement, but the family is not implemented in Tryll.

Note the two distinct reasons. Kokoro and Kitten are blocked by licensing — they cannot be enabled without changing what Tryll redistributes. VITS, Matcha, and ZipVoice are simply not implemented; nothing in principle prevents their eSpeak-free bundles from being added later.

Why eSpeak-dependent families are unavailable

Sherpa-ONNX performs grapheme-to-phoneme conversion for several families with eSpeak NG, which is licensed GPLv3-or-later. Linking it into tryll_server.exe would make the server a combined work under GPLv3 — and because the server payload is bundled inside your game build, that obligation would follow you, your studio, and your publisher.

Tryll therefore builds Sherpa-ONNX with eSpeak NG compiled out entirely. The shipped sherpa-onnx-c-api.dll contains no eSpeak code and the payload contains no espeak-ng-data. The upside is that nothing in the Tryll server imposes a copyleft obligation on your game. The cost is that any voice whose front-end calls eSpeak cannot run, regardless of how the model's own weights are licensed.

A permissive model license is not enough

KittenTTS is the classic trap: its weights are Apache-2.0, which looks unrestricted, but the model is trained on eSpeak phoneme IDs and Sherpa-ONNX's Kitten loader requires a populated espeak-ng-data directory. The permissive headline license does not make the runtime dependency go away. The same reasoning excludes Kokoro, whose loader also hard-requires that directory.

What happens if you try an unsupported family

A models.json variant whose tts_family is anything other than supertonic or pocket fails when the model is loaded — that is, at CreateAgent time for a graph referencing it, not silently at startup.

tts_family value Result
supertonic, pocket Loads normally
vits, piper, kokoro, or empty Rejected with an explicit eSpeak/GPLv3 message
Any other value (kitten, matcha, zipvoice, typos) Rejected as an unknown family

Both cases surface as agent-creation failures; see Error Codes. An empty tts_family is rejected rather than defaulting, because the historical default was VITS.

Adding your own voice

You can add TTS entries to models.json — either a HuggingFace repo or a local path — using the schema in Model Management. The constraint is the family, not the source: the bundle must be a Supertonic or Pocket bundle, with the tts_files key set that family expects.

Bundles for both families are published under csukuangfj on HuggingFace alongside the rest of the Sherpa-ONNX model zoo. A bundle from any other family will be rejected at load time even though the catalog entry itself parses.