Skip to content
SPECIFICTION
Get listed

Google Cloud Text-to-Speech

cloud.google.com

From $4/mo Free trial WebiOSAndroidAPI
Visit cloud.google.com
Spec score 9.2 #8 of 233 in Text-to-speech tools

What it is

Google Cloud Text-to-Speech is an API that converts text or SSML into audio data for applications. It offers more than 380 voices across more than 75 languages and variants, with REST or gRPC integration and output formats including MP3, LINEAR16, OGG_OPUS, MULAW, and ALAW. Gemini-TTS supports single- or multispeaker speech guided by natural-language prompts for style, accent, pace, tone, and emotional expression. Chirp 3 instant custom voice can create a personalized voice model from as little as 10 seconds of audio and is available in more than 30 locales. Streaming synthesis supports real-time conversations, while long-audio synthesis accepts up to 1 million bytes of input. Speech controls include pitch, speaking rate, and volume gain. Billing counts input characters, including spaces, newlines, and SSML tags except the mark tag, and billing must be enabled. Standard and WaveNet voices cost US$4 per 1 million characters, with 0 to 4 million characters free. Other voice models have different rates and free-use limits. New customers can receive $300 in Google Cloud credits.

Spec sheet

What it does
Cloud Text-to-Speech converts text or SSML input into natural human speech audio.Read 15 Oct 2026 · docs.cloud.google.com
Purpose
The API converts text into natural-sounding speech and can accept text or SSML input to produce audio data.Read 4 Oct 2026 · cloud.google.com
Platforms
web,apiRead 20 Sep 2026 · cloud.google.com
Free plan
NoRead 20 Sep 2026 · cloud.google.com
Integrations
GetApp lists integrations including Cloud Vision API, YouTube, Unity and Studio.Read 15 Oct 2026 · getapp.com
Security
Google Cloud states that data is encrypted in transit and at rest with audited access to encryption keys.Read 15 Oct 2026 · cloud.google.com
Compliance
Google Cloud describes ISO 27001/27017/27018/27701, SOC 1/2/3, PCI DSS and FedRAMP offerings plus GDPR and HIPAA alignment.Read 15 Oct 2026 · cloud.google.com
License
The public Google API protocol definition is licensed under Apache License 2.0; the hosted synthesis service itself is proprietary.Read 15 Oct 2026 · github.com
API access
YesRead 20 Sep 2026 · cloud.google.com
API positioning
The product is an API powered by Google AI technologies for generating natural-sounding speech.Read 15 Oct 2026 · cloud.google.com
Audio formats
The service can return formats including MP3, Linear16 and OGG Opus.Read 15 Oct 2026 · getapp.com
Authentication
Client libraries support Application Default Credentials for programmatic authentication.Read 15 Oct 2026 · docs.cloud.google.com
Billing limit
Billing counts input characters including spaces, newlines, and SSML tags except the mark tag, and billing must be enabled to use the service.Read 4 Oct 2026 · cloud.google.com
Chirp 3 HD
Chirp 3 HD voices provide low-latency streaming, expressive intonation and human disfluencies for conversational agents.Read 15 Oct 2026 · cloud.google.com
Commercial use
YesRead 20 Sep 2026 · cloud.google.com
Company founded
Google's official history dates the company's beginnings to 1998.Read 15 Oct 2026 · about.google
Company headquarters
Google's official history identifies the Googleplex in Mountain View, California as its headquarters.Read 15 Oct 2026 · about.google
Company legal name
Google's terms identify the provider as Google LLC.Read 15 Oct 2026 · about.google
Company size
LinkedIn lists Google as a public software-development company with 10,001-plus employees and headquarters in Mountain View, California.Read 15 Oct 2026 · linkedin.com
Custom voices
Chirp 3 instant custom voice can create personalized voice models with as little as 10 seconds of audio input and is available in more than 30 locales.Read 4 Oct 2026 · cloud.google.com
Developer APIs
Google documents REST v1 and v1beta1 APIs, gRPC APIs, authentication guidance and client libraries.Read 15 Oct 2026 · docs.cloud.google.com
Documentation depth
Documentation includes product basics, quickstarts, SSML tutorials, voice catalogs, API references, authentication, samples and release notes.Read 15 Oct 2026 · docs.cloud.google.com
Export formats
MP3,LINEAR16,OGG_OPUS,MULAW,ALAWRead 20 Sep 2026 · cloud.google.com
Gemini-TTS
Gemini-TTS supports single- and multispeaker synthesis with natural-language control of style, accent, pace, tone and emotion.Read 15 Oct 2026 · cloud.google.com
GetApp feature ratings
GetApp reports AI voices 4.8/5, custom voices 5.0/5, multilingual support 5.0/5, API 5.0/5 and SSML 5.0/5 from small review subsets.Read 15 Oct 2026 · getapp.com
Instant custom voice
Instant custom voice can create a personalized voice model from as little as 10 seconds of audio in more than 30 locales.Read 15 Oct 2026 · cloud.google.com
Interfaces and formats
Applications can integrate through REST or gRPC requests, and output formats include MP3, Linear16, and OGG Opus.Read 4 Oct 2026 · cloud.google.com
Latest release
On December 30, 2025, Chirp 3 HD added preview synthesis for Chinese Hong Kong (yue-HK).Read 15 Oct 2026 · docs.cloud.google.com
Limit: Capterra reviewer usability
A Capterra reviewer says reaching the correct page is difficult because there is no simple standalone application.Read 15 Oct 2026 · capterra.com
Limit: high-volume cost
G2 reviewers repeatedly warn that costs rise quickly for large audio volumes and premium voices.Read 15 Oct 2026 · g2.com
Limit: setup complexity
G2 reviewers report that console setup, authentication and SSML require technical expertise and have a learning curve.Read 15 Oct 2026 · g2.com
Long audio
Long-audio synthesis asynchronously accepts up to 1 million bytes of input.Read 15 Oct 2026 · cloud.google.com
Long audio limit
Long audio synthesis supports up to 1 million bytes of input.Read 4 Oct 2026 · cloud.google.com
Notable news
On April 16, 2026, Google announced Gemini 3.1 Flash TTS preview with 200-plus audio tags and support across more than 70 languages.Read 15 Oct 2026 · cloud.google.com
Praise: Capterra reviewers
Capterra reviewers praise easy text conversion, versatile voice generation and strong perceived value.Read 15 Oct 2026 · capterra.com
Praise: integration
G2 reviewers repeatedly describe the API and Google Cloud integrations as straightforward and scalable.Read 15 Oct 2026 · g2.com
Praise: voice quality
G2 reviewers repeatedly praise natural, lifelike voices, especially Neural2, WaveNet and newer models.Read 15 Oct 2026 · g2.com
Rating on Capterra
4.7 out of 5 from 12 reviewsRead 15 Oct 2026 · capterra.com
Rating on G2
4.4 out of 5 from 273 reviewsRead 15 Oct 2026 · g2.com
Rating on GetApp
4.7 out of 5 from 12 reviewsRead 15 Oct 2026 · getapp.com
SDKs
Official client-library examples cover Node.js, PHP, Python, Java, Go, C#, Ruby and other languages.Read 15 Oct 2026 · docs.cloud.google.com
Security and compliance
Google Cloud says it maps security, privacy, and compliance controls to standards and undergoes independent verification through certifications, attestations, and audit reports.Read 4 Oct 2026 · cloud.google.com
Speech controls
The API supports pitch adjustment up to 20 semitones, speaking rates up to four times faster or slower than normal, and volume gain from -96 dB to +16 dB.Read 4 Oct 2026 · cloud.google.com
SSML
SSML controls pauses, number and date formatting, pronunciation and other delivery details.Read 15 Oct 2026 · docs.cloud.google.com
Streaming
The product supports streaming audio synthesis for low-latency, real-time conversations.Read 4 Oct 2026 · cloud.google.com
Streaming synthesis
The feature list includes streaming audio synthesis for ultra-low-latency conversations.Read 15 Oct 2026 · cloud.google.com
Support and training
GetApp lists email/help desk, forums, knowledge base, phone, chat, live online training, webinars, documentation and videos.Read 15 Oct 2026 · getapp.com
Support resources
The documentation page provides guides, API references, quotas and limits, release notes, and a getting-support resource.Read 4 Oct 2026 · docs.cloud.google.com
Trial credits
New customers can receive $300 in Google Cloud credits to try Text-to-Speech and other Google Cloud products.Read 4 Oct 2026 · cloud.google.com
Use cases
Google describes use cases including contact-center voicebots, voice generation in devices, and accessible electronic program guides.Read 4 Oct 2026 · cloud.google.com
Voice catalog
Google lists more than 380 voices across more than 75 languages and regional variants.Read 15 Oct 2026 · cloud.google.com
Voice cloning
YesRead 20 Sep 2026 · cloud.google.com
Voice selection
Google lists more than 380 voices across more than 75 languages and variants.Read 4 Oct 2026 · cloud.google.com
Voice tuning
Pitch can be adjusted by up to plus or minus 20 semitones, speaking rate can be changed up to four times faster or slower, and gain ranges from -96 dB to +16 dB.Read 15 Oct 2026 · cloud.google.com

Plans and prices

Standard voices $4 a month per 1 million characters after 4 million free characters 0–4 million characters free Read 15 Oct 2026 · cloud.google.com
WaveNet voices $4 a month per 1 million characters after 4 million free characters 0–4 million characters free Read 15 Oct 2026 · cloud.google.com
Neural2 voices $16 a month per 1 million characters after 1 million free characters 0–1 million characters free Read 15 Oct 2026 · cloud.google.com
Polyglot voices $16 a month Price after free usage limit is reached 1 million characters free usage Read 20 Sep 2026 · cloud.google.com
Chirp 3: HD voices $30 a month per 1 million characters after 1 million free characters 0–1 million characters free · low-latency streaming · expressive conversational voices Read 15 Oct 2026 · cloud.google.com
Studio voices $160 a month per 1 million characters after 1 million free characters 0–1 million characters free Read 15 Oct 2026 · cloud.google.com
Instant custom voice $60 a month per 1 million characters no free usage · personalized voice models Read 15 Oct 2026 · cloud.google.com
Gemini 2.5 Flash TTS and Gemini 2.5 Flash-Lite Preview TTS Price not published Input: $0.50 per 1 million text tokens · Output: $10.00 per 1 million audio tokens · No free usage limit listed Read 20 Sep 2026 · cloud.google.com
Gemini 3.1 Flash TTS Price not published Preview · Input: $1.00 per 1 million text tokens · Output: $20.00 per 1 million audio tokens · No free usage limit listed Read 20 Sep 2026 · cloud.google.com
Gemini 2.5 Flash TTS / Gemini 2.5 Flash-Lite Preview TTS Price not published Input tokens: $0.50 per 1 million text tokens; output tokens: $10.00 per 1 million audio tokens No free usage limit Read 4 Oct 2026 · cloud.google.com
Gemini 2.5 Pro TTS Price not published Input tokens: $1.00 per 1 million text tokens; output tokens: $20.00 per 1 million audio tokens No free usage limit Read 4 Oct 2026 · cloud.google.com
Gemini 3.1 Flash TTS (Preview) Price not published Input tokens: $1.00 per 1 million text tokens; output tokens: $20.00 per 1 million audio tokens No free usage limit Read 4 Oct 2026 · cloud.google.com

Every Google Cloud Text-to-Speech plan and price →

Strong points and limits

Strong points

  • More than 380 voices across 75+ languages and variants
  • Accepts text or SSML input
  • Supports streaming audio synthesis
  • New customers can receive $300 in Google Cloud credits

Limits

  • Billing must be enabled to use the service
  • Billing counts spaces, newlines, and most SSML tags
  • Pricing and free-use limits vary by voice model

Who it suits

It suits developers adding speech output to applications through an API, including contact-center voicebots, device voice generation, and accessible electronic program guides. Its voice controls and model choices also suit projects needing different speaking styles or formats.

Google Cloud Text-to-Speech in depth

Read the full write-up

Google Cloud Text-to-Speech is an API for turning text or SSML into speech audio. It is best suited to developers adding spoken output to products, rather than people looking for a personal read-aloud app. Its breadth of voices and controls is a strength, but model-specific billing calls for careful cost planning.

Overview

The service offers more than 380 voices across over 75 languages and variants, with REST and gRPC integration. That range suits apps serving multiple markets, while streaming synthesis can support low-latency conversations such as contact-center voicebots. It is an infrastructure choice, not a finished voice application.

Long-audio synthesis accepts up to 1 million bytes of input. Billing must be enabled, and character-based charges count spaces, newlines, and SSML tags, except the mark tag. Those rules make both text formatting and voice-model selection relevant to operating cost.

Key features

Voice choice and control

Gemini-TTS supports single- or multispeaker speech, with natural-language prompts to guide style, accent, pace, tone, and emotional expression. This is useful when a product needs directed delivery rather than a neutral readout, though Gemini uses token-based pricing rather than the per-character rates of other voice families.

Chirp 3 instant custom voice can create a personalized voice model from as little as 10 seconds of audio and is available in more than 30 locales. The API also supports pitch adjustment up to 20 semitones, speaking rates up to four times slower or faster than normal, and volume gain from -96 dB to +16 dB. These controls give developers room to tune output, while custom voice and higher-end models have different costs that should be assessed before choosing them.

Streaming and output

Streaming synthesis is intended for real-time, low-latency interactions. Audio can be exported as MP3, Linear16, OGG Opus, MULAW, or ALAW, giving applications several output options. The combination suits products that need generated speech in an existing workflow, including voice-enabled devices and accessible electronic program guides.

Security and support

Google Cloud maps security, privacy, and compliance controls to standards and undergoes independent verification through certifications, attestations, and audit reports. Documentation includes guides, API references, quotas and limits, release notes, and a support resource, which helps teams working through integration and operational constraints.

Pricing

Pricing varies by voice family. The Standard and WaveNet plans each show free, with 0 to 4 million characters free and then US$0.000004 per character (US$4 per 1 million characters). These are the clearest fit for high-volume projects that can use those voices and want a stated per-character rate.

Chirp 3: HD voices show free, with 0 to 1 million characters free and a rate of US$0.00003 per character (US$30 per 1 million characters). Studio voices also show free and include 0 to 1 million characters free, but cost US$0.00016 per character (US$160 per 1 million characters). They may suit projects prioritizing those voice families, but the higher per-character rates make usage estimates especially important.

Neural2 and Polyglot each show $16 a month and include 1 million characters of free usage; the price after that limit is reached is not given. Instant custom voice shows free, costs US$0.00006 per character (US$60 per 1 million characters), and has no free usage limit. A new Google Cloud customer can also receive $300 in credits to try Text-to-Speech and other Google Cloud products.

Gemini pricing is token-based, with no free usage limit stated. Gemini 2.5 Flash TTS and Gemini 2.5 Flash-Lite Preview TTS are priced at $0.50 per 1 million text tokens and $10.00 per 1 million audio tokens. Gemini 2.5 Pro TTS is $1.00 per 1 million text tokens and $20.00 per 1 million audio tokens; Gemini 3.1 Flash TTS is a preview at the same input and output rates. Gemini 3.8 Flash TTS (Preview), through Dec 31, 2026, is $0.50 per 1 million text tokens and $9.00 per 1 million audio tokens. These model-specific rates make direct cost comparison with character-priced voices dependent on the workload.

Platforms

The service is available through an API and on the web. Applications can connect through REST or gRPC requests. Commercial use is supported, and the API provides access to the speech-generation capabilities and export formats.

Who it's for

Google Cloud Text-to-Speech is a strong fit for developers building voicebots, speech-enabled devices, or accessible program guides, particularly when they need broad language coverage, streaming, or fine control of speech. Teams that want personalized voices can consider Chirp 3 instant custom voice, while teams with unpredictable usage should compare the chosen model's billing basis and free allowance against expected traffic. It is not the natural choice for someone seeking a standalone app to listen to documents.

Pros and cons

Pro: More than 380 voices across more than 75 languages and variants give applications broad voice and market coverage.

Pro: Streaming, multiple audio formats, and substantial pitch, rate, and volume controls support varied product requirements.

Pro: Gemini-TTS prompting and Chirp 3 instant custom voice offer more delivery direction and personalization than a fixed voice selection alone.

Con: Pricing spans character and token billing, with materially different rates by model; estimating costs requires knowing the intended voice family and usage.

Con: Billing must be enabled, and character charges include whitespace and most SSML tags, which can make formatted input count toward usage.

Con: This is an API-centered service, so people seeking a ready-to-use reading app should choose a different kind of product.

Alternatives

AI voice generators, text-to-speech tools, and game voice generators are broader categories to explore if you want to compare options by use case.

Typecast is worth considering if you want an Android or iOS option alongside web and API access; its free plan includes three projects, 3,000 lifetime credits, 1 GB media storage, and 720p video exports.

Fish Audio may suit users who want self-hosted or desktop-platform options; its free tier includes 8,000 monthly credits and up to seven minutes of generation, with a 500-character per-generation limit.

Kokoro TTS is a free alternative with an 82-million-parameter model and Apache-licensed weights.

LOVO AI offers a free plan limited to five projects; its Pro+ plan includes 20 voice-generation hours per month and unlimited projects.

NaturalReader is an option for readers who want apps across mobile, desktop, browser extension, and web platforms; its Lite plan is $79 a year for one user and 1 million MP3 characters per month.

Speechify may fit personal listening across mobile, desktop, and web; its free plan includes 10 voices and a maximum speed of 1.5x.

Acapela Group is a paid alternative available across mobile, desktop, web, API, and self-hosted platforms, with a yearly subscription priced at €99 a year excluding VAT.

CereProc is another paid option across API, apps, desktop, web, and self-hosted platforms; its CereVoice Cloud 1 Million Credit plan costs £12.99 a month.

Verdict

Choose Google Cloud Text-to-Speech if you are building a product that needs programmable speech, broad voice coverage, streaming, or configurable delivery. Its main advantage is the range of integration and voice options in one API; its main drawback is that costs depend sharply on model family and billing unit. For a personal text-reading app or a project that needs a single predictable price, look elsewhere.

Verdict

Google Cloud Text-to-Speech offers a wide range of voices, controls, and integration options through an API. Review the per-model rates and character-based billing rules before estimating usage costs.

Where Specifiction places it

Alternatives to Google Cloud Text-to-SpeechAll alternatives →