Bark

Bark is Suno's open-source transformer-based text-to-audio model. Unlike conventional TTS, it generates audio directly from text prompts with no phoneme intermediate — so the same prompt language controls speech, music, background noise, sound effects, and nonverbal sounds like [laughs], [sighs], [gasps], hesitations (...), emphasis via CAPS, and ♪ sung lyrics ♪. It is multilingual, ships large and small checkpoints (both on Hugging Face), is available in Transformers from 4.31.0, and comes with a voice prompt library and long-form generation notebooks. MIT-licensed since May 2023 with checkpoints explicitly available for commercial use. Suno flags it as research-grade: being fully generative, it can deviate from prompts unexpectedly.

1. Install from git (verified from README — do NOT pip install bark, that is an unrelated package):

```bash

pip install git+https://github.com/suno-ai/bark.git

```

or git clone https://github.com/suno-ai/bark && cd bark && pip install .

1. Generate audio:

```python

from bark import SAMPLE_RATE, generate_audio, preload_models

preload_models() # downloads checkpoints on first run

audio_array = generate_audio("Hello, my name is Suno. [laughs]")

```

1. Use a voice prompt for consistency: generate_audio(text, history_prompt="v2/en_speaker_6") — browse the voice prompt library linked in the README.
2. Alternatively use the Transformers integration (transformers>=4.31.0) or the Hugging Face Space demo.