Instantly transform text into natural-sounding voiceovers. Explore 25+ voices spanning American and British accents using our AI TTS Platformca
No generations yet
Your generated speech files will appear here
TTS Chat converts your written text into natural speech using a neural network model. Select a voice, choose your format, and generate audio in seconds.
Powered by Kokoro TTS, a neural network model that produces human-like speech with accurate intonation, rhythm, and emphasis across 25+ voices.
Choose from American and British English accents, both male and female. Each voice has been fine-tuned for natural pronunciation and clarity.
Audio begins playing as soon as the model generates the first waveform — no need to wait for the full file. Skip, seek, and re-listen instantly.
Download your generated speech as MP3, WAV, Opus, AAC, or FLAC. High-quality audio suitable for podcasts, voiceovers, accessibility, and more.
25 distinct voices from 4 accent groups: American Female (11), American Male (9), British Female (4), and British Male (1). Each with unique tonal qualities.
No API keys or accounts required. All generation happens server-side and no text or audio data is stored — your content stays yours.
Every voice listed here was generated and curated from the Kokoro TTS model. New voices are added as they become available.
Everything you need to know about using the TTS Chat tool.
Kokoro is an open-weight text-to-speech model that produces natural, expressive speech. It runs on a dedicated server and supports fine-grained voice control through embedding files.
There are 25 voices across four groups: American Female (Alloy, Aoede, Bella, Heart, Jessica, Kore, Nicole, Nova, River, Sarah, Sky), American Male (Adam, Echo, Eric, Fenrir, Liam, Michael, Onyx, Puck, Santa), British Female (Alice, Emma, Isabella, Lily), and British Male (Daniel).
Text input is limited to 2,000 characters per generation, which typically yields 3–5 minutes of speech depending on voice and pacing. For longer content, generate in segments.
You can export in MP3 (good balance of size and quality), WAV (uncompressed), Opus (best compression), AAC (Apple-compatible), and FLAC (lossless). MP3 is recommended for most use cases.
Yes. The audio output is yours to use for any purpose — commercial projects, podcasts, voiceovers, accessibility features, or personal use. No attribution required.
There are no hard rate limits for the lab tool. Each request is capped at 2,000 characters. If you need higher volume or production SLAs, please get in touch.