🎙️ Parler-TTS Mini with ParaSpeechCaps

Generate expressive speech with rich style control using our Parler-TTS model finetuned on ParaSpeechCaps. Control various aspects of speech including:

  • Speaker characteristics (pitch, clarity, etc.)
  • Emotional qualities
  • Speaking style and rhythm

Choose between two models:

  • Full Model: Trained on complete ParaSpeechCaps dataset
  • Base Model: Trained only on human-annotated ParaSpeechCaps-Base
Model

Choose between the full model or base-only model

0 3
1 5
0 50
ASR Model

ASR model used for quality assessment

Examples
Style Description Text to Synthesize
Pages: