THE COST OF BUILDING WITH AI

Audio model APIs: speech & music pricing

Browse supported speech and music-generation variants, their billing units and output contracts. Audio generation is available; transcription is not currently offered.

Try in Playground Calculate your cost

Published by Tokely · Reviewed 2026-10-09

THE SHORT ANSWER

Tokely supports selected audio-generation variants with model-specific billing units. Speech and music are separate contracts; use the endpoint, voice settings and format documented for the model. Speech-to-text is not currently supported.

Published rates for enabled audio models

Published Tokely rates for each supported variant. Units and settings differ: this is not a same-output quality or cheapest-provider ranking. Open a model for full parameters and input requirements.
Model	Rate & unit	Variant settings	Input contract
Gemini 2.5 Pro TTSGoogle · Audio	$0.6 / maximum reservationMetered generation	voice: Kore	prompt
Gemini 3.1 Flash TTSGoogle · Audio	$0.0075 / generation creditMeasured speech usage	voice: Kore	prompt
Suno V6Suno · Audio	$0.09 / taskOne bounded generation	See model parameters	prompt

Model developers in this category

Google ↗

Gemini 2.5 Pro TTS, Gemini 3.1 Flash TTS

Suno ↗

Suno V6

Connect with the supported contract

Open a model page for the exact ID and request example. Use a Tokely key and funded balance. Read the audio cost guide or start with the quickstart.

START WITH YOUR OWN WORKLOAD

Know the price.
Then make the call.

Check the model, compare the rate, and connect with a Tokely key.

Read the quickstart ↗Explore all models →