Skip to main content
POST
/
synthesis
Perform speech synthesis
curl --request POST \
  --url https://api.example.com/synthesis \
  --header 'Content-Type: application/json' \
  --data '
{
  "accent_phrases": [
    {
      "moras": [
        {
          "text": "<string>",
          "vowel": "<string>",
          "vowel_length": 123,
          "pitch": 123,
          "consonant": "<string>",
          "consonant_length": 123
        }
      ],
      "accent": 123,
      "is_interrogative": false
    }
  ],
  "speedScale": 123,
  "pitchScale": 123,
  "intonationScale": 123,
  "volumeScale": 123,
  "prePhonemeLength": 123,
  "postPhonemeLength": 123,
  "outputSamplingRate": 123,
  "outputStereo": true,
  "pauseLength": 123,
  "pauseLengthScale": 1,
  "kana": "<string>"
}
'
"<string>"

Query Parameters

speaker
integer
required
enable_interrogative_upspeak
boolean
default:true

Automatically adjust sentence-final intonation when the text is interrogative

core_version
string

Body

application/json

Query for speech synthesis.

accent_phrases
AccentPhrase · object[]
required

List of accent phrases

speedScale
number
required

Overall speech speed

pitchScale
number
required

Overall pitch

intonationScale
number
required

Overall intonation

volumeScale
number
required

Overall volume

prePhonemeLength
number
required

Silence duration before the audio

postPhonemeLength
number
required

Silence duration after the audio

outputSamplingRate
integer
required

Output sampling rate of the audio data

outputStereo
boolean
required

Whether to output audio data in stereo

pauseLength
number | null

Silence duration for punctuation etc. Ignored when null. Default value is null.

pauseLengthScale
number
default:1

Silence duration for punctuation etc. (multiplier). Default value is 1.

kana
string

[Read-only] Text in AquesTalk-style notation. Ignored as a query for speech synthesis.

Response

Successful Response

The response is of type file.

Last modified on May 27, 2026