Speech Synthesis

Speech synthesis is one of the most important parts of building conversational AI. We provide you with the tool to synthesize as human-like speech as possible.

There are mainly two ways for you to use our text-to-speech engine.

The first one is to use say or sayText directives in a node. For example:

#sayText("Welcome to United Postal Service! How can I help you?"); #say("dont_understand");

Another way to try out our TTS is through CLI with only one command:

dasha tts synthesize --speaker Kate "Hello, how are you?" -o speech.wav

Available speakers

Custom voices

We can create your own custom voice. If interested, email us at support@dasha.ai

Notes

Emotional Speech Synthesis (Beta) and Voice cloning (Beta) are depreciated and are no longer availe.

Found a mistake? Let us know.

Enroll in beta

Request invite to our private Beta program for developers to join the waitlist. No spam, we promise.