Best Free Text to Speech Online: Limits Compared (2026)

There is no single best free text to speech tool. The right choice depends on whether you need a downloadable MP3, a more natural voice, commercial rights or a long script. As of October 2026, free options range from browser voices that need no account to monthly character or credit allowances on paid platforms. The table below compares the limits.

Free text to speech limits compared

These allowances were checked in October 2026. Vendors change their plans often, so confirm the live page before you commit.

Read the table first, then the catches section. The fine print on downloads and commercial rights often matters more than the headline allowance, because a generous free plan can still be unusable for a project that needs an MP3 or a client license.

| Tool | Free allowance | Notable limits | |---|---|---| | ElevenLabs | 10,000 credits per month, about 10 minutes of speech | No commercial license listed on Free | | NaturalReader | Unlimited use of Free Voices; 20,000 characters per day with Lite Voices; 4,000 shared characters per day with Plus and Cloned Voices | Free Voices cannot produce downloadable MP3; help article dated 13 July 2026 | | Speechify | Listening app with speeds up to 1.5x and 10 robotic sounding voices | Text to speech features only; Premium is $29 a month | | Amazon Polly | 5 million Standard characters per month; 1 million Neural characters for the first 12 months | Long-Form (500,000) and Generative (100,000) allowances last 12 months | | Azure AI Speech (F0) | 0.5 million Neural characters per month | Paid prices did not render on the page checked | | Gemini API | Gemini 3.8 Flash TTS, 3.8 Flash-Lite TTS, 3.1 Flash TTS Preview and 2.5 Flash Preview TTS are free of charge | The page says free-tier use may be used for product improvement | | OndaLabs | $0 for 2,000 credits per month (about 2 minutes) | Onda Lite model; commercial use starts on Starter |

What "free" actually measures

Each tool counts usage in a different unit, so the allowances cannot be compared directly.

ElevenLabs quotes credits, with approximate minutes.
NaturalReader quotes characters per day.
Amazon Polly quotes characters per month.
Gemini quotes tokens.

Counting rules also change the result. Azure counts every billable character, including punctuation, spaces and markup, and each Chinese character counts as two. A script that is short in English can use far more of an Azure allowance in another language.

The catches: downloads, commercial rights and data use

Free plans usually differ from paid plans in three places:

Downloads: NaturalReader's help page says its free plan cannot convert to MP3, and its Free Voices cannot produce downloadable MP3 files at all.
Commercial rights: ElevenLabs lists a commercial license from Starter upward, not on its Free plan. OndaLabs includes commercial use on Starter and above.
Data use: The Gemini API pricing page says its free tier may be used for product improvement, so read that before you send private scripts.

OpenAI's usage policies also require a clear disclosure that a TTS voice is AI-generated, whatever plan you use.

Free tiers for developers

Developers have more options, but each one has its own rules:

Amazon Polly: New AWS customers receive up to $200 in AWS Free Tier credits, starting July 15, 2025.
Azure AI Speech: The F0 tier includes 0.5 million Neural characters per month.
Google Cloud Text-to-Speech: Has a free tier. Check its pricing page for the current allowances.
Gemini API: The four TTS models above are free of charge on the Standard tier. The page says paid Standard prices for the two 3.8 models double on 1 January 2027.
OpenAI: Its pricing page lists no free TTS tier. Paid prices include $15 per 1 million characters for tts-1.

Free without an account: browser voices

The Web Speech API's speechSynthesis interface is free and needs no account. According to MDN, it uses voices supplied by the operating system, so the available voices and their quality vary by device and browser. This is a good option for quick listening or prototypes, but it gives you less control over the voice. For a first test of a script, that is often enough to hear whether the wording reads well before you spend credits on a paid service.

Free and open source: run it yourself

Kokoro-82M is an open-weight model with 82 million parameters, released under the Apache-2.0 license. You can deploy it yourself at no licensing cost. Its v1.0 release, published 27 January 2025, supports 8 languages and 54 voices. Its model card warns that sites with "kokoro" in the domain, such as kokorottsai_com and kokorotts_net, are not affiliated with the model or its author. Running it yourself means you also handle hosting and setup.

How to choose: a short checklist

Do you need an MP3 or other download? Check the download rules before you sign up.
Will the audio be used commercially? Confirm the plan lists a commercial license.
How many minutes a month do you need? Convert the allowance into your own unit, since each tool counts differently.
Which languages and voices do you need? Test them with your actual text.
How long is each script? ElevenLabs lists per-request limits of 10,000 characters for Eleven v4, 5,000 for v3, 10,000 for Multilingual v2 and 40,000 for Flash v2.5.

If you are new to these tools, what text to speech is gives the basics.

Where OndaLabs' free plan fits

OndaLabs' Free plan costs $0 and includes 2,000 credits a month, about 2 minutes, on Onda Lite. It also includes expressive tags, two-speaker dialogue, speech to text and 5 saved clips. Commercial use is not part of the Free plan. Starter costs $9 and includes 170,000 credits, about 189 minutes, with access to Onda Flash at 1.5 credits per character and commercial use.

Be direct about the trade-off. Two minutes a month is a trial-sized allowance. ElevenLabs lists 10,000 credits, about 10 minutes, on its Free plan. Those units differ by model, so the two numbers are not a ranking. If you need more, OndaLabs top-ups are $5 for 100,000 credits. You can try the free plan in the text to speech studio or compare plan limits on the pricing page.

Frequently asked questions

What is the best free text to speech online?

No single tool wins for every need. Browser voices are free and need no account, but their quality depends on the device. Vendor free plans suit short tests, and their limits and rights differ. Match the tool to your needs for downloads, commercial use and script length.

Is there free text to speech with no character limit?

Not as a general promise. NaturalReader's help page describes unlimited use of its Free Voices, which are basic system voices. Its Lite and Plus voice allowances are daily caps, and the other free plans in this comparison all have monthly or daily allowances.

Can I download MP3 audio from free text to speech?

It depends on the tool. NaturalReader's Free Voices cannot produce downloadable MP3 files, and its help page says the free plan cannot convert to MP3. Check each tool's download rules on its current plan page before you rely on one.

Can I use free text to speech audio commercially?

Often not on the free plan. ElevenLabs lists no commercial license on its Free plan, and OndaLabs includes commercial use on Starter and above. Read the license terms of the plan you use before you publish audio for a client or a product.

What is the best free open-source text to speech?

One open-weight option is Kokoro-82M, which is released under the Apache-2.0 license and can run on your own hardware. "Best" depends on your languages, voices and hardware, so test a short passage before you choose. Make sure you download the model from the official model card.