BASE MODEL · TTS

Bark

Generative audio model that can laugh, sigh and emote.

Open on Hugging Face ↗ Hear 1 voice built on it

Voices built on Bark

Speaker embedding

Pip ★ 4.1

EN English · British Child · Energetic · Expressive Character · Gaming · 5.2K downloads

Try voice

Bark · Speaker embedding Commercial OK

Technical details

Architecture: GPT-style text-to-audio transformer
Type: TTS
Size: 5 GB
Runs on: ~12 GB VRAM
Languages: English and 12 more
Recommended settings: text_temp=0.7 · waveform_temp=0.7
Version: v0 · updated 1 yr ago
Source: Hugging Face
License: MIT · Commercial OK
Original repository: huggingface.co/suno/bark ↗

Bark is developed and owned by its original authors. timbre only indexes it.

COMPARE Compare