TTS (Premium)

Article summary

Did you find this summary helpful?

Thank you for your feedback

Available in Classic and VPC

Synthesize speech by taking in text to convert and parameters like tone, speed, and emotion.

Request

The following describes the request format for the endpoint. The request format is as follows:

Method	URI
POST	/tts

Request headers

For headers common to all CLOVA Voice APIs, see Common CLOVA Voice headers.

Request body

The following describes the request body.

Field	Type	Required	Description
`speaker`	String	Required	Voice type to use for speech synthesis For information about each voice, see List of speech synthesis voices
`text`	String	Required	Text to be converted to speech Only support UTF-8-encoded text Synthesize speech up to 2000 characters Text in symbols or parentheses is not converted
`volume`	Integer	Optional	Speech volume `-5` - `5` (default: `0`) `-5`: Synthesize 0.5x smaller `0`: Synthesize to normal volume `5`: Synthesize 1.5x larger
`speed`	Integer	Optional	Speech speed `-5` - `5` (default: `0`) `-5`: Synthesize 2x faster `0`: Synthesize to normal speed `5`: Synthesize 0.5x slower
`pitch`	Integer	Optional	Speech pitch `-5` - `5` (default: `0`) `-5`: Synthesize 1.2x higher `0`: Synthesize to normal pitch `5`: Synthesize 0.8x lower
`emotion`	Integer	Optional	Emotion level of speech Supported voices: `nara` \| `vara` \| `vmikyung` \| `vdain` \| `vyuna` \| `vgoeun` \| `vdaeseong` For information about each voice, see List of speech synthesis voices `0` - `3` (default: `0`) `0`: neutral `1`: sad `2`: happy `3`: angry (`nara` not supported)
`emotion-strength`	Integer	Optional	Emotion intensity of speech Supported voices: `vara` \| `vmikyung` \| `vdain` \| `vyuna` \| `vgoeun` \| `vdaeseong` For information about each voice, see List of speech synthesis voices `0` - `2` (default: `1`) `0`: weak `1`: normal `2`: strong
`format`	String	Optional	Speech file format `mp3` (default) \| `wav`
`sampling-rate`	Integer	Optional	Sampling rate of speech Only supported for `wav` format `8000` \| `16000` \| `24000` (default)\| `48000` Exceptionally, `mijin` supports the `16000` rate for `mp3` format
`alpha`	Integer	Optional	Tone `-5` - `5` (default: `0`) If higher than `0`: high tone If lower than `0`: low tone
`end-pitch`	Integer	Optional	End-pitch processing of speech Supported voices: `clara` \| `matt` \| `meimei` \| `liangliang` \| `chiahua` \| `kuanlin` \| `carmen` \| `jose` \| all voices starting with `d-` (<e.g.> `dara`) For information about each voice, see List of speech synthesis voices `-5` - `5` (default: `0`) If higher than `0`: high end-pitch If lower than `0`: low end-pitch

List of speech synthesis voices

The following describes in detail the types of voices that will be used for speech synthesis.

Field	Name	Language	Description
`nara`	Ara	Korean	Female
`nara_call`	Ara (agent)	Korean	Female
`nminyoung`	Minyoung	Korean	Female
`nyejin`	Yejin	Korean	Female
`mijin`	Mijin	Korean	Female
`jinho`	Jinho	Korean	Male
`clara`	Clara	English	Female
`matt`	Matt	English	Male
`shinji`	Shinji	Japanese	Male
`meimei`	Meimei	Chinese	Female
`liangliang`	Liangliang	Chinese	Male
`jose`	Jose	Spanish	Male
`carmen`	Carmen	Spanish	Female
`nminsang`	Minsang	Korean	Male
`nsinu`	Sinu	Korean	Male
`nhajun`	Hajun	Korean	Child (male)
`ndain`	Dain	Korean	Child (female)
`njiyun`	Jiyun	Korean	Female
`nsujin`	Sujin	Korean	Female
`njinho`	Jinho	Korean	Male
`njihun`	Jihun	Korean	Male
`njooahn`	Jooahn	Korean	Male
`nseonghoon`	Seonghoon	Korean	Male
`njihwan`	Jihwan	Korean	Male
`nsiyoon`	Siyoon	Korean	Male
`ngaram`	Garam	Korean	Child (female)
`ntomoko`	Tomoko	Japanese	Female
`nnaomi`	Naomi	Japanese	Female
`dnaomi_joyful`	Naomi (happy)	Japanese	Female
`dnaomi_formal`	Naomi (news)	Japanese	Female
`driko`	Riko	Japanese	Female
`deriko`	Eriko	Japanese	Female
`nsayuri`	Sayuri	Japanese	Female
`ngoeun`	Goeun	Korean	Female
`neunyoung`	Eunyoung	Korean	Female
`nsunkyung`	Sunkyung	Korean	Female
`nyujin`	Yujin	Korean	Female
`ntaejin`	Taejin	Korean	Male
`nyoungil`	Youngil	Korean	Male
`nseungpyo`	Seungpyo	Korean	Male
`nwontak`	Wontak	Korean	Male
`dara_ang`	Ara (angry)	Korean	Female
`nsunhee`	Sunhee	Korean	Female
`nminseo`	Minseo	Korean	Female
`njiwon`	Jiwon	Korean	Female
`nbora`	Bora	Korean	Female
`njonghyun`	Jonghyun	Korean	Male
`njoonyoung`	Joonyoung	Korean	Male
`njaewook`	Jaewook	Korean	Male
`danna`	Anna	English	Female
`djoey`	Joey	English	Female
`dhajime`	Hajime	Japanese	Male
`ddaiki`	Daiki	Japanese	Male
`dayumu`	Ayumu	Japanese	Male
`dmio`	Mio	Japanese	Female
`chiahua`	Chiahua	Taiwanese	Female
`kuanlin`	Kuanlin	Taiwanese	Male
`nes_c_hyeri`	Hyeri	Korean	Female
`nes_c_sohyun`	Sohyun	Korean	Female
`nes_c_mikyung`	Mikyung	Korean	Female
`nes_c_kihyo`	Kihyo	Korean	Male
`ntiffany`	Kiseo	Korean	Female
`napple`	Neulbom	Korean	Female
`njangj`	Dream	Korean	Female
`noyj`	Bomdal	Korean	Female
`neunseo`	Eunseo	Korean	Female
`nheera`	Heera	Korean	Female
`nyoungmi`	Youngmi	Korean	Female
`nnarae`	Narae	Korean	Female
`nyeji`	Yeji	Korean	Female
`nyuna`	Yuna	Korean	Female
`nkyunglee`	Kyunglee	Korean	Female
`nminjeong`	Minjeong	Korean	Female
`nihyun`	Ihyun	Korean	Female
`nraewon`	Raewon	Korean	Male
`nkyuwon`	Kyuwon	Korean	Male
`nkitae`	Kitae	Korean	Male
`neunwoo`	Eunwoo	Korean	Male
`nkyungtae`	Kyungtae	Korean	Male
`nwoosik`	Woosik	Korean	Male
`vara`	Ara (Pro)	Korean	Female
`vmikyung`	Mikyung (Pro)	Korean	Female
`vdain`	Dain (Pro)	Korean	Female
`vyuna`	Yuna (Pro)	Korean	Female
`vhyeri`	Hyeri (Pro)	Korean	Female
`dara-danna`	Ara & Anna	Korean + English (U.S.)	Female
`dsinu-matt`	Sinu & Matt	Korean + English (U.S.)	Male
`nsabina`	Witch Sabina	Korean	Female
`nmammon`	Demon Mammon	Korean	Male
`nmeow`	Meow	Korean	Child (female)
`nwoof`	Woof	Korean	Child (male)
`nreview`	Review Park	Korean	Male
`nyounghwa`	Movie Jeong	Korean	Female
`nmovie`	Movie Choi	Korean	Male
`nsangdo`	Sangdo	Korean	Male
`nshasha`	Shasha	Korean	Female
`nian`	Ian	Korean	Male
`ndonghyun`	Donghyun	Korean	Male
`vian`	Ian (Pro)	Korean	Male
`vdonghyun`	Donghyun (Pro)	Korean	Male
`dsayuri`	Sayuri	Japanese	Female
`dtomoko`	Tomoko	Japanese	Female
`dnaomi`	Naomi	Japanese	Female
`vgoeun`	Goeun (Pro)	Korean	Female
`vdaeseong`	Daeseong (Pro)	Korean	Male
`ngyeongjun`	Gyeongjun	Korean	Male
`ndaeseong`	Daeseong	Korean	Male
`njonghyeok`	Jonghyeok	Korean	Male

Request example

The following is a sample request.

curl --location --request POST 'https://naveropenapi.apigw.ntruss.com/tts-premium/v1/tts' \
--header 'X-NCP-APIGW-API-KEY-ID: {Client ID issued when registering the app}' \
--header 'X-NCP-APIGW-API-KEY: {Client secret issued when registering the app}' \
--header 'Content-Type: application/x-www-form-urlencoded' \
--data-urlencode 'speaker=nara' \
--data-urlencode 'text=Hello' \
--data-urlencode 'volume=0' \
--data-urlencode 'speed=-1' \
--data-urlencode 'pitch=1' \
--data-urlencode 'emotion=2' \
--data-urlencode 'emotion-strength=1' \
--data-urlencode 'format=wav' \
--data-urlencode 'sampling-rate=8000' \
--data-urlencode 'alpha=0' \
--data-urlencode 'end-pitch=0'

Response

The following describes the response format.

Response body

The following describes the response body.

Field	Type	Required	Description
Responded TTS audio	Binary	-	Binary voice data in MP3 or WAV format

Response status codes

For response status codes common to all CLOVA Voice APIs, see Common CLOVA Voice response status codes.

Response example

The following is a sample example.

{Binary voice data in MP3 or WAV format}

Was this article helpful?

What's Next

CLOVA Voice examples

Table of contents

Request
Response