gnani/timbre-v1.5

Timbre v1.5

Timbre v1.5 is a text-to-speech model trained with full supervised fine-tuning (SFT).

Model Details

• Version: 1.5
• Training: Full SFT
• Architecture: Transformer-based TTS model with AudioVAE

Usage

# Usage example coming soon

Training Details

This model was trained using full supervised fine-tuning on audio datasets.

License

Apache 2.0