Bark

Transformer text-to-audio model that speaks, sings, laughs, and makes sound effects.

Bark, from Suno, is a transformer-based text-to-audio model that generates highly realistic multilingual speech as well as nonverbal sounds like laughter, sighing, music, and noise. Prompted with bracketed cues and speaker presets, it can produce expressive narration and ambient audio rather than purely clean read-aloud TTS.