Text to speech in 100+ languages
Every Auditext voice can speak every supported language, so one character can narrate your English, Spanish and Japanese versions. These guides cover the details that make each language sound right.
- Spanish Text to SpeechSpanish is one of the easiest languages for a speech engine to read, because its spelling maps closely to sound.
- French Text to SpeechFrench is hard for older speech engines because so much is written but not said: silent final consonants, liaisons ("les amis" sounds like "lé-z-ami") and elision.
- German Text to SpeechGerman spelling is regular, but long compound nouns (Donaudampfschifffahrt) and word stress are where synthetic voices can stumble.
- Japanese Text to SpeechJapanese text to speech has two hard problems: choosing the right reading for each kanji (日 can be "hi", "nichi" or "ka") and applying pitch accent, the rise and fall that separates 箸 (hashi, chopsticks) from 橋 (hashi, bridge).
- British Accent Text to SpeechA British accent is more than swapping a few vowels.
Languages in the studio menu
The language menu in the text to speech studio lists the most requested languages and accents: English (US), English (British accent), Spanish (Spain), Spanish (Latin America), French, German, Italian, Portuguese (Brazil), Dutch, Japanese, Korean, Chinese (Mandarin), Hindi, Arabic. For anything else, choose “Detect from text” — the voices handle more than 100 languages.
For listening practice without limits, Read Aloud uses the voices installed on your device, which usually include a dozen or more languages.
More from the studio
- AI text to speechType or upload, pick one of 30 voices, download MP3.
- AI voice generatorBrowse voices or describe one and let the casting assistant pick.
- Text to MP3Render text to an MP3 at the bitrate you choose.
- PDF to audioPull the text out of a PDF and listen or export.
- Read aloudUnlimited reading with word highlighting, using your device voices.
- Voiceover script writerDraft or tighten a script timed to your video length.
- Words to minutesHow long your script or speech will take to say.
- SSML editorBuild pauses, emphasis and prosody tags and convert them.
- How TTS worksStep through what happens between text and sound.