Suno's Bark Generates Speech in 13 Languages and Allows Commercial Use
Summary
Suno's Bark is a transformer text-to-audio model that generates speech in 13 languages plus music, laughter and sound effects, and it carries an MIT license permitting commercial use. The full model needs about 12GB of VRAM and holds roughly 13 seconds per generation; a smaller variant fits 8GB.
Key Points
- Suno says Bark offers more than 100 speaker presets and does not support custom voice cloning.
- Bark converts text directly to audio without phonemes, so prompts can carry tags like [laughs] and [MAN].
- Suno notes that SUNO_OFFLOAD_CPU with small models lets Bark run on GPUs with about 2GB of VRAM.