Supported languages

The real-time API accepts ISO-639-1 language codes (en, es, zh, …). Regional tags are accepted (en-US, pt-BR) — only the primary subtag is used.

---

Input (sourceLanguage)

Languages you can speak. Pass auto to let the ASR detect the spoken language; the detected code is returned on every transcript frame as detected_language.

Code	Language
auto	Auto-detect
zh	Chinese (Mandarin)
yue	Cantonese
en	English
fr	French
de	German
it	Italian
ja	Japanese
ko	Korean
pt	Portuguese
ru	Russian
es	Spanish

---

Output — translated text (targetLanguage with audioOutputEnabled=false)

Languages you can receive as text transcripts.

Code	Language
ar	Arabic
zh	Chinese
cs	Czech
nl	Dutch
en	English
fr	French
de	German
hi	Hindi
id	Indonesian
it	Italian
ja	Japanese
ko	Korean
pl	Polish
pt	Portuguese
ru	Russian
es	Spanish
th	Thai
tr	Turkish
vi	Vietnamese

---

Output — translated speech (targetLanguage with audioOutputEnabled=true)

Languages you can receive as translated audio. Voices are synthesized via Piper. Where a language has both a male and female voice in the Piper catalog, you can select with voiceType; languages marked with a single voice synthesize the same voice regardless of voiceType.

Code	Language	Voices
ar	Arabic	single
bg	Bulgarian	single
ca	Catalan	male, female
cs	Czech	single
cy	Welsh	male, female
da	Danish	single
de	German	male, female
el	Greek	single
en	English	male, female
es	Spanish	male, female
eu	Basque	male, female
fa	Persian (Farsi)	male, female
fi	Finnish	single
fr	French	male, female
hi	Hindi	male, female
hu	Hungarian	male, female
id	Indonesian	single
is	Icelandic	male, female
it	Italian	male, female
ka	Georgian	single
kk	Kazakh	male, female
lb	Luxembourgish	single
lv	Latvian	single
ml	Malayalam	male, female
ne	Nepali	single
nl	Dutch	male, female
no	Norwegian	single
pl	Polish	male, female
pt	Portuguese	male, female
ro	Romanian	single
ru	Russian	male, female
sk	Slovak	single
sl	Slovenian	single
sq	Albanian	single
sr	Serbian	single
sv	Swedish	single
sw	Swahili	single
te	Telugu	male, female
tr	Turkish	single
uk	Ukrainian	male, female
ur	Urdu	single
vi	Vietnamese	single
zh	Chinese (Mandarin)	single

Japanese, Korean, Cantonese, and Thai do not currently have a Piper voice. Use them with audioOutputEnabled=false to get text transcripts.

Requesting an unsupported targetLanguage with audioOutputEnabled=true returns an error frame. Pass audioOutputEnabled=false to get transcripts in any of the text-output languages above.