PRICING & CREDITS
Gemini TTS pricing is based on your model and submitted character count. See the current application rates and review the live estimate before every take.
Every tier includes the same voice features.
$9.90/month
1,800 credits every month
≈ 2.7 hours of Flash speech
≈ 150k characters at the Flash rate
$19.90/month
3,800 credits every month
≈ 5.8 hours of Flash speech
≈ 317k characters at the Flash rate
$39.90/month
7,800 credits every month
≈ 11.9 hours of Flash speech
≈ 650k characters at the Flash rate
Speech duration is an estimate from recent samples. Actual duration varies with the script and pace; the studio shows the character-based charge before generation.
For expressive narration and two-speaker dialogue.
1.2credits per started 100 characters
A lower credit rate for readings and script revisions.
0.8credits per started 100 characters
Character counts include every dialogue turn. The estimate updates as you edit your script.
Speech requests use started blocks of 100 characters. A script with 100 characters uses one block, while a script with 101 characters starts a second block. This is a character calculation, not a word count or a charge based on the finished audio duration. The selected model determines the credit rate applied to those blocks.
For example, a 1,000-character script currently estimates 12.0 credits with Flash and 8.0 with Flash-Lite. Check the live estimate after editing text or changing models. In two-speaker mode, the submitted dialogue text counts toward the same request; each speaker does not receive a separate character allowance.
Gemini TTS pricing applies to each new generation request, so allow for the additional takes you choose to create. Start with a representative paragraph that includes important names and a typical sentence. Once you are happy with the voice and delivery, use those settings as the starting point for a longer segment. Each request supports up to 5,000 characters.
Browsing the voice library, listening to its existing previews, and applying a text template do not submit a speech request. A template fills editable controls in the studio. Review its text and settings before choosing Generate speech. This makes it possible to prepare the draft and see its estimate before committing to a take.
New accounts receive 20 registration credits, valid for 30 days from registration. The gift is a limited starting balance, not an unlimited free generation plan. Monthly plans and one-time credit packs are available above.
The Gemini TTS pricing shown here belongs to this independent application. It is not a quotation for Google’s direct API service, and an external API account does not supply credits to this studio. Use the model pages to compare workflows and the estimate in the generator to understand the current request.
The studio shows the estimate and your available balance after sign-in. If the request needs more credits, it offers a link back to this page. You can shorten the script or compare the other model’s estimate before making another attempt. Changing the model alone does not generate audio.
A word can contain a few letters or many characters. Two scripts with the same number of words can therefore use different numbers of billing blocks. Punctuation and the actual text you submit also matter. Use the character counter and live estimate in the studio instead of converting a document’s word count into credits yourself.
Include the task identifier from your library, the model selected, and the error message you saw. Mention whether the issue happened before submission, during generation, or while downloading the result. Never send passwords or access tokens. This context helps distinguish a balance question from a playback or delivery problem.
Yes. When a task completes, play the audio in the studio and download the WAV file you want to keep. Completed takes are also available through your library. Review a failed or unfinished task there before submitting a duplicate request.
