Providers

Parler-TTS (Hugging Face)

TTS 3 models no stock voices open weights only data from provider API synced 2026-11-18

Open-source TTS project from Hugging Face. Parler-TTS generates speech whose speaker style (gender, pitch, speaking rate, recording quality) is controlled by a natural-language text description; the Mini (880M) and Large (2.3B) v1 checkpoints were trained on 45k hours of audiobook data and released under Apache-2.0 with their datasets and training code.

Text-to-speech 3

CompareModelLangsFeaturesPriceStatus
parler-tts/parler-tts-large-v1parler-tts-large-v1 1 0/8 –
parler-tts/parler-tts-mini-v1parler-tts-mini-v1 1 0/8 –
parler-tts/parler_tts_mini_v0.1parler_tts_mini_v0.1 1 0/8 –
Esc

↑↓ move↵ open