This document contains current API restrictions and usage limits on use of
Cloud Text-to-Speech. This page will be updated to reflect any changes to these
restrictions and usage limits. We reserve the right to change these limits. To
keep track of your content and request totals, see the documentation about
[monitoring your API usage](https://cloud.google.com/apis/docs/monitoring).

You can use the audio data files you create using Cloud Text-to-Speech in
your applications or media in compliance with the
[Google Cloud Platform Terms of Service](https://cloud.google.com/terms/) including compliance
with all applicable law.

You can request an increase to the
[requests limit](https://docs.cloud.google.com/text-to-speech/quotas.md.txt#requests_limit) from the
[Google Cloud console](https://console.cloud.google.com/). The
[content limits](https://docs.cloud.google.com/text-to-speech/quotas.md.txt#content_limits) cannot be increased.
For more information, see [Request a quota adjustment](https://docs.cloud.google.com/docs/quotas/view-manage.md.txt).

After submitting your request, Google may contact you for more
information, and inform you whether your request is approved or denied.

## Content limits

Content to Cloud TTS is provided as text data, either as
raw strings or SSML-formatted data.

The API contains the following limits on the size of this content (and
are subject to change):

| Speech Synthesis Limit | Value |
|---|---|
| Total bytes per request | 5,000 |

> [!NOTE]
> **Note:** In some locales (for example ja-JP), a single character is made up of multiple bytes.

## Requests limit

The current API usage limits for Cloud TTS are as follows
(and are subject to change):

| Type of Limit | Usage Limit |
|---|---|
| Chirp3RequestsPerMinutePerProject | 200 |
| ChirpVoiceCloningRequestsPerMinutePerProject | 30 |
| ConcurrentStreamingSessionsPerProject | 100 |
| GenerateVoiceCloningKeyRequestsPerMinutePerProject | 10 |
| LongAudioSynthesisQueryRequestsPerMinutePerProject | 100 |
| LongAudioSynthesisRequestsPerMinutePerProject | 100 |
| Neural2RequestsPerMinutePerProject | 1,000 |
| PolyglotRequestsPerMinutePerProject | 1,000 |
| RequestsPerMinutePerProject | 1,000 |
| StudioRequestsPerMinutePerProject | 500 |

> [!NOTE]
> **Note:** The `RequestsPerMinutePerProject` quota applies to all Cloud TTS voices that don't have a dedicated quota listed in this table.

## Gemini-TTS Usage Limits

Starting with Gemini 3.1 Flash TTS, Gemini-TTS models support
dynamic throughput-based usage limits. See a detailed explanation of [consumption options](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/deploy/consumption-options).

| Model name | Consumption options |
|---|---|
| gemini-3.1-flash-tts-preview | [Standard PayGo](https://docs.cloud.google.com/vertex-ai/generative-ai/docs/standard-paygo) |

The API usage limits for earlier models are as follows:

| Model name | Usage limit (QPM^1^) |
|---|---|
| gemini-2.5-flash-tts | 150^2^ |
| gemini-2.5-pro-tts | 125^2^ |

^1^Queries Per Minute

^2^The usage limits stated are default maximum values and they are subject to change. The effective quota limits may vary by project, and may be increased upon request.