mlsanityData to 14 Oct 2026

Deterministic Prompting for Speaker-Stable Low-Resource Greek TTS

arXiv 2609.10022 ↗First posted 9 Sep 2026cs.SD cs.CL cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Modern TTS systems approach human quality for high-resource languages but degrade when clean speech data is scarce. Modern Greek exemplifies this, lacking the curated corpora behind state-of-the-art synthesis. We propose a data curation recipe that transforms audiobook recordings into TTS-ready data via WhisperX alignment and filtering. Then we fine-tune Parler-TTS (880M), a prompt-based multilingual model whose pre-training encodes phonetic priors transferable to Greek. During development, we find that LLM-generated style prompts introduce speaker drift at inference. Replacing them with deterministic prompts resolves this, and a speaker-specific LoRA stage trained on 3.5 h of single-speaker data anchors identity while updating ~5% of parameters. Our system achieves WER 10.7% (2.9 above the ASR floor), MOS-I 4.00 (vs. 4.36 human speech), and near-human speaker consistency (MOS-C 4.24 vs. 4.30), showing that robust single-speaker Greek TTS is achievable with limited curated data.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 22 of its 30 references that are in the map. First posted on arXiv 9 Sep 2026.

Builds on

22 of 30 references.

References · 22

  1. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  2. 2022Scaling Instruction-Finetuned Language Modelscited by 3,024
  3. 2020HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesiscited by 1,154
  4. 2017Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictionscited by 804
  5. 2023Neural Codec Language Models are Zero-Shot Text to Speech Synthesizerscited by 752
  6. 2020FastSpeech 2: Fast and High-Quality End-to-End Text to Speechcited by 717
  7. 2020ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verificationcited by 628
  8. 2021Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speechcited by 561
  9. 2017Tacotron: Towards End-to-End Speech Synthesiscited by 490
  10. 2023Simple and Controllable Music Generationcited by 488
  11. 2021Grad-TTS: A Diffusion Probabilistic Model for Text-to-Speechcited by 447
  12. 2023High-Fidelity Audio Compression with Improved RVQGANcited by 431
10 more
  1. 2021SpeechBrain: A General-Purpose Speech Toolkitcited by 406
  2. 2023WhisperX: Time-Accurate Speech Transcription of Long-Form Audiocited by 326
  3. 2020Glow-TTS: A Generative Flow for Text-to-Speech via Monotonic Alignment Searchcited by 273
  4. 2024Natural language guidance of high-fidelity text-to-speech with synthetic annotationscited by 67
  5. 2019CSS10: A Collection of Single Speaker Speech Datasets for 10 Languagescited by 54
  6. 2024Zero-shot Voice Conversion with Diffusion Transformerscited by 43
  7. 2024An Initial Investigation of Language Adaptation for TTS Systems under Low-resource Scenarioscited by 4
  8. 2022Regotron: Regularizing the Tacotron2 architecture via monotonic alignment losscited by 3
  9. 2024A multilingual training strategy for low resource Text to Speechcited by 3
  10. 2024Leveraging Parameter-Efficient Transfer Learning for Multi-Lingual Text-to-Speech Adaptationcited by 1

Built on by

Not cited yet.