FreyaTTS Technical Report
2026-07-10 • Computation and Language
Computation and Language
AI summaryⓘ
The authors present Freya-TTS, a compact Turkish text-to-speech model that generates speech efficiently without using phonemes or complex tokenizers. It uses a special type of diffusion transformer working on a compressed audio representation, enabling high-quality 48 kHz sound output. Their approach improves voice consistency and handles short utterances well, achieving low error rates on speech benchmarks with fewer parameters than larger models. Freya-TTS runs very fast on both GPUs and CPUs, making it suitable for devices with limited resources. The authors also provide the model and code openly under an Apache license.
text-to-speechDiffusion Transformernon-autoregressiveAudioVAElatent spacegrapheme-to-phonemeword error ratecharacter error ratereal-time factorTurkish language processing
Authors
Ahmet Erdem Pamuk, Ömer Yentür, Ahmet Tunga Bayrak, Yavuz Alp Sencer Öztürk, Mustafa Yavuz
Abstract
We introduce Freya-TTS, a compact, tokenizer-free, Turkish-first text-to-speech model designed for highly reliable and efficient conversational synthesis. Freya-TTS is a 183.2M-parameter non-autoregressive conditional flow-matching Diffusion Transformer (DiT) that operates in the frozen continuous latent space of AudioVAE2 (16 kHz encode, 48 kHz decode), allowing the model to focus its capacity on text-to-latent mapping while inheriting high-quality 48 kHz reconstruction. We advance the framework along three key dimensions: (1) rule-free end-to-end modeling from a 92-symbol Turkish character vocabulary without a phonemizer, grapheme-to-phoneme frontend, or discrete speech tokenizer; (2) non-autoregressive parallel denoising, which predicts the entire latent sequence simultaneously over a predicted duration; and (3) a production-oriented two-stage post-training recipe consisting of single-speaker voice locking and short-utterance coverage, improving speaker consistency and robustness on short inputs. On the Freya-TR-Eval benchmark, Freya-TTS achieves a band-matched word error rate (WER) of 8.0% and character error rate (CER) of 3.0%, outperforming substantially larger open-source systems while using a fraction of their parameters. The model achieves a real-time factor of 0.11 on consumer GPUs and runs faster than real time on a laptop CPU, making it well suited for resource-constrained edge deployment. We release the model weights, training and inference code, and evaluation benchmark under the Apache-2.0 license.