26 natural voices
Choose from 26 voices: 6 OpenAI voices plus 20 ElevenLabs Turbo v2.5 voices, each previewable before you spend credits.
Text to Speech
Turn text into natural speech with 26 voices across OpenAI and ElevenLabs. 32 languages, up to 40,000 characters, 10 credits per 1,000.
Text to Speech
Turn text into natural speech with 26 voices across OpenAI and ElevenLabs. 32 languages, up to 40,000 characters, 10 credits per 1,000.
Paste text and hear it read back in a natural voice. 26 voices across OpenAI and ElevenLabs Turbo v2.5, 32 languages, documents up to 40,000 characters, at 10 credits per 1,000 characters.
Krater pricing · AI model directory · AI Image Generator and Photo Editor
Same line, three different voices
Hello, this is a test of the Krater text to speech engine. Type a line, pick a voice, and hear it read back in seconds.
ElevenLabs Turbo v2.5, Aria · 10 credits
Hello, this is a test of the Krater text to speech engine. Type a line, pick a voice, and hear it read back in seconds.
ElevenLabs Turbo v2.5, Roger · 10 credits
Hello, this is a test of the Krater text to speech engine. Type a line, pick a voice, and hear it read back in seconds.
ElevenLabs Turbo v2.5, River · 10 credits
Voices, languages and length that cover real work
32 languages
Synthesize speech in 32 languages, from English and Spanish to Arabic, Hindi and Japanese, with an auto-detect option.
Documents up to 40K characters
ElevenLabs jobs accept up to 40,000 characters, roughly 40 minutes of audio, so articles and scripts go in one run.
Priced by text size
10 credits per 1,000 characters with a 10 credit minimum. A 2,000 character script costs 20 credits, shown before it runs.
Adjustable speed
Slow narration down for courses or speed it up for drafts, without changing the voice.
Custom voice IDs
Paste an ElevenLabs voice ID to use your own cloned or library voices through the same /tts command.
Popular models
From text to audio in four steps
Type /tts and paste the text
The script goes right after the command. Short clips and long documents both work.
Pick voice and language
Choose one of 26 voices and any of 32 languages, or leave language on auto-detect.
Set the speed
Adjust playback speed for narration, drafts or accessibility listening.
Play or download the audio
The synthesized clip arrives in the chat at 10 credits per 1,000 characters.
Who turns text into speech
Video voiceovers
Narrate clips and product videos without recording a take.
Podcasts and narration
Draft episode intros, ads and readings in a consistent voice.
Accessibility
Turn written content into audio for listeners who prefer or need it.
Courses and training
Produce lesson narration in any of 32 languages.
Articles to audio
Convert long reads into listenable form, up to about 40 minutes per job.
Copy a prompt, run it as /tts
- /tts Welcome back to the channel. Today we are looking at three tools that save an hour a day.
- /tts Chapter one. It was the coldest morning of the year when the letter finally arrived.
- /tts In this lesson we cover how photosynthesis turns light into energy, step by step.
- /tts <paste your article here to listen to it instead of reading>
When a different tool fits better
No voice cloning from scratch
The 26 built-in voices cover most needs and custom ElevenLabs voice IDs work, but Krater does not clone a voice from an uploaded sample today.
Very long books need splitting
One job tops out at 40,000 characters on ElevenLabs (4,096 on OpenAI voices). Full audiobooks need chapter-by-chapter runs.
One-off free use
There is no free plan for TTS. Pro starts at $20 per month. For a single clip, a free reader is enough.
Text to speech questions
Is the text to speech tool free to use?
Not fully. /tts runs on credits at 10 credits per 1,000 characters. Pro is $20 per month and includes monthly credits.
How many voices and languages are there?
26 voices total: 6 from OpenAI and 20 from ElevenLabs Turbo v2.5. The ElevenLabs voices cover 32 languages with an auto-detect option.
How much text can I convert at once?
Up to 40,000 characters per job on the ElevenLabs voices, about 40 minutes of audio, and 4,096 characters on the OpenAI voices.
Can I use my own voice?
If you have an ElevenLabs voice ID, paste it into the voice options and /tts uses it. Uploading an audio sample to clone a voice inside Krater is not supported today.
Related tools and models
- AI Video Generator
Put the narration over generated footage, all in one chat.
- Audio to Text
Transcribe recordings back into editable text.
- AI Sound Effects
Generate backing tracks and effects for the same project.
- AI model directory
Every model on Krater with credit pricing and capabilities.