Flux TTS Voices & Languages
The Flux TTS voice catalog. Every voice is a flux-{voice}-{language} model string served on /v2/speak over both the streaming WebSocket and batch REST transports.
Flux TTS voices use the model string format flux-{voice}-{language} (e.g. flux-haley-en). The same voice catalog is served on both /v2/speak transports — the streaming WebSocket and batch REST.
Selecting a voice
Pass the model string on connection:
Every voice handles general conversational synthesis. Use the tables below to shortlist, then hear the voices live at talk.deepgram.com before you build.
Featured voices
Start here. The featured voices are the strongest all-rounders in the catalog — the most versatile across customer service, IVR, narration, and casual chat. Browse More voices below for additional accents, ages, and characters when you’re matching a specific brand or audience.
More voices
The rest of the catalog. Every voice here handles general conversational synthesis; the featured voices above are the strongest all-rounders.
Dutch
Dutch (Flemish)
French
German
German (Swiss)
Hindi
Italian
Japanese
Portuguese (Brazil)
Portuguese (Portugal)
Russian
Spanish
Bengali — Early Access
Bulgarian — Early Access
Catalan — Early Access
Chinese (Mandarin) — Early Access
Chinese (Taiwanese Mandarin) — Early Access
Chinese (Cantonese) — Early Access
Croatian — Early Access
Czech — Early Access
Danish — Early Access
Estonian — Early Access
Finnish — Early Access
Greek — Early Access
Gujarati — Early Access
Hungarian — Early Access
Indonesian — Early Access
Kannada — Early Access
Korean — Early Access
Latvian — Early Access
Lithuanian — Early Access
Malay — Early Access
Marathi — Early Access
Norwegian — Early Access
Polish — Early Access
Romanian — Early Access
Slovak — Early Access
Slovenian — Early Access
Swedish — Early Access
Tagalog — Early Access
Tamil — Early Access
Telugu — Early Access
Turkish — Early Access
Ukrainian — Early Access
Urdu — Early Access
Vietnamese — Early Access
Languages
Flux TTS voices speak English — with accents spanning American, British, Irish, Australian, Indian, Singaporean, and Filipino English — and 41 other languages:
Bengali (bn), Bulgarian (bg), Catalan (ca), Chinese (zh), Croatian (hr), Czech (cs), Danish (da), Dutch (nl), Estonian (et), Finnish (fi), French (fr), German (de), Greek (el), Gujarati (gu), Hindi (hi), Hungarian (hu), Indonesian (id), Italian (it), Japanese (ja), Kannada (kn), Korean (ko), Latvian (lv), Lithuanian (lt), Malay (ms), Marathi (mr), Norwegian (no), Polish (pl), Portuguese (pt), Romanian (ro), Russian (ru), Slovak (sk), Slovenian (sl), Spanish (es), Swedish (sv), Tagalog (tl), Tamil (ta), Telugu (te), Turkish (tr), Ukrainian (uk), Urdu (ur), Vietnamese (vi).
Languages marked Early Access do not have text normalization yet: numbers, dates and abbreviations are read as written, so write them out in words. Text normalization is available for English, German, Dutch, French, Hindi, Italian, Japanese, Portuguese, Russian and Spanish.
The voices in these 41 languages are available on api.deepgram.com, api.eu.deepgram.com and api.in.deepgram.com; they are not yet available on the Australia endpoint (api.au.deepgram.com).
Related resources
- Getting Started (Streaming) — connect and synthesize with a voice
- Getting Started (Batch) — one-shot synthesis
- Feature Overview
- Migrating from Aura to Flux TTS