🎵 Audio Models

Speech recognition, text-to-speech, and AI music generation

OpenAI

Best Transcription

OpenAI's state-of-the-art speech recognition. 99 languages, near-human accuracy, and diarization support.

$0.006 / minute 25 MB audio GA

ElevenLabs

Most Natural TTS

The most natural-sounding TTS available. Voice cloning, 29 languages, emotional control, and real-time streaming.

$0.30 / 1k characters — GA

Suno

Best Music Gen

Generate full studio-quality songs from text descriptions. Vocals, instruments, and mixing from a single prompt.

$8 / month (500 credits) — GA

Udio

Best Audio Quality

Udio's latest model produces extremely high-fidelity music with precise genre control and remixing capabilities.

$10 / month — GA

Suno AI

Open Source

Open-source text-to-audio model that generates speech with realistic emotion, music, and environmental sounds.

Free (self-hosted) — GA

Select a Model

Choose a model from the list to start chatting

Token usage0

Ask a question to see prompt and answer tokens.

🤖

Select a model from the left to begin