Kokoro
Small open-weight text-to-speech model
Kokoro is an 82 million parameter open-weight text-to-speech model. It is small enough to run on modest hardware and is released under the Apache 2.0 license.
More in Text-to-speech
Featured
ElevenLabs
Text-to-speech
Text-to-speech API and studio with many voices
Amazon Polly
Text-to-speech
AWS text-to-speech service
Azure AI Speech Text to Speech
Text-to-speech
Neural voices from Microsoft Azure
Cartesia
Text-to-speech
Low-latency speech models for realtime agents
Coqui TTS (Idiap fork)
Text-to-speech
Maintained fork of the Coqui TTS toolkit
Fish Audio
Text-to-speech
Text-to-speech and voice cloning platform