Model reference · open weights
Freya-TTS is an open-weight audio or speech model from freyavoice. AxForge deploys and operates it for you on dedicated EU-owned hardware — with the licence handled where one is required.
Available as managed deployment — configured and operated for you on dedicated EU hardware, quoted per deployment.
What it is
| Released by | freyavoice |
|---|---|
| Type | Audio & music |
| Task | Text→speech |
| Parameters (lead) | 183M |
| Runs with | freyatts |
| Released | 2026-07-07 |
| Popularity | 7k downloads / month |
| Licence | Open weights |
About
FreyaTTS-small is a 183M-parameter Turkish text-to-speech model, and the open-source member of the FreyaTTS family. It is tokenizer-free at the character level (92 symbols, no phonemizer or G2P) and generates speech with a non-autoregressive conditional flow-matching DiT in the frozen AudioVAE2 latent space (25 Hz, 64-dim latents, 16 kHz encode / 48 kHz decode). Output is 48 kHz mono.
FreyaTTS-small (this model) is our compact, Apache-2.0, self-hostable model, released in full: weights, inference code, and training pipeline.
FreyaTTS-large is our production model, serving Turkish voice agents at Freya (YC S25) with higher naturalness and expressivity. It is available commercially rather than as open weights. For access, contact us at dev@freyavoice.ai.
The model id freyavoice/freya-tts predates this naming and is unchanged; it refers to FreyaTTS-small.
from freyatts import FreyaTTS
tts = FreyaTTS.from_pretrained("freyavoice/freya-tts", device="cuda")
wav = tts.synthesize("Merhaba, size nasıl yardımcı olabilirim?") # np.float32, 48 kHz
tts.save_wav(wav, "output.wav")
Clone freyavoiceai/FreyaTTS and install requirements.txt.
On Freya-TR-Eval: WER 8.0% / CER 3.0%, ranking 3rd of 7 among open sub-1B Turkish TTS models, ahead of XTTS-v2 (11.1% WER) and F5-TTS (24.3% WER).
FreyaTTS-small is described in our technical report, arXiv:2607.09530:
@misc{pamuk2026freyattscompacttokenizerfreeflowmatching,
title={FreyaTTS: A Compact Tokenizer-Free Flow-Matching Transformer for Turkish-First Speech Synthesis},
author={Ahmet Erdem Pamuk and Ömer Yentür and Ahmet Tunga Bayrak and Yavuz Alp Sencer Öztürk and Mustafa Yavuz},
year={2026},
eprint={2607.09530},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2607.09530},
}
From the published model card. Full card on the HuggingFace links in the sidebar.
How it works
Using it via the API
Once AxForge deploys freya-tts for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (freya-tts below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/audio/transcriptions \ -H "Authorization: Bearer $AXFORGE_API_KEY" \ -F model="freya-tts" -F file=@audio.mp3
Create an account — your API key is available in the console. 3M free tokens every 30 days with every new account.