speech

create a streaming session

mint a one-shot websocket session for real-time streaming synthesis. set audio_format here to negotiate opus, pcm, mulaw, alaw, or mp3; omit it for raw 24 khz mono signed 16-bit little-endian pcm. connect only to the returned ws_url, append ?token=<token>, and send a json synthesis frame. binary messages contain the negotiated audio stream and are followed by a terminal {"type":"done"} json frame. the returned token is single-use and expires in 300 seconds. on the socket, errors arrive as {"error": true, "code", "recoverable"} frames with no done after them: INVALID_TEXT (nothing to speak, not charged) and INVALID_MESSAGE (the frame was not a json object) are recoverable and the socket stays open.

POST

tts

ws-connect

curl --request POST \
  --url https://silk-api.rumik.ai/v1/tts/ws-connect \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "mulberry",
  "text": "streaming in real time."
}
'

{
  "ws_url": "wss://silk-api.rumik.ai/ws/tts",
  "token": "<string>",
  "request_id": "<string>",
  "expires_in": 300
}

{
  "error": "<string>",
  "code": "<string>"
}

{
  "error": "<string>",
  "code": "<string>"
}

{
  "error": "<string>",
  "code": "<string>"
}

{
  "error": "text: Field required",
  "code": "VALIDATION_ERROR",
  "detail": [
    {
      "type": "missing",
      "loc": [
        "body",
        "text"
      ],
      "msg": "Field required"
    }
  ]
}

{
  "error": "<string>",
  "code": "<string>"
}

{
  "error": "<string>",
  "code": "<string>"
}

Authorizations

Authorization

string

header

required

your rumik api key, e.g. rk_live_.... create one in the rumik dashboard.

Body

application/json

text

string

required

text to synthesize. sent again in the websocket synthesis frame.

Maximum string length: 2000

model

enum<string>

default:muga

which model to use.

Available options:

muga,

mulberry

audio_format

enum<string>

optional output format for this one-shot session. omit the field for raw 24 khz mono signed 16-bit little-endian pcm. explicit null and unsupported values return 400 UNSUPPORTED_AUDIO_FORMAT.

Available options:

opus,

pcm,

mulaw,

alaw,

mp3

Response

a one-shot websocket session.

ws_url

string<uri>

websocket url to connect to.

Example:

"wss://silk-api.rumik.ai/ws/tts"

token

string

one-shot token. pass as ?token=<token> when connecting. single-use; expires after expires_in seconds.

request_id

string

id of this request. the socket's done and cancelled frames carry a request_id too.

expires_in

integer

seconds the token stays redeemable (300).

Example:

300

curl --request POST \
  --url https://silk-api.rumik.ai/v1/tts/ws-connect \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "mulberry",
  "text": "streaming in real time."
}
'

{
  "ws_url": "wss://silk-api.rumik.ai/ws/tts",
  "token": "<string>",
  "request_id": "<string>",
  "expires_in": 300
}

{
  "error": "<string>",
  "code": "<string>"
}

{
  "error": "<string>",
  "code": "<string>"
}

{
  "error": "<string>",
  "code": "<string>"
}

{
  "error": "text: Field required",
  "code": "VALIDATION_ERROR",
  "detail": [
    {
      "type": "missing",
      "loc": [
        "body",
        "text"
      ],
      "msg": "Field required"
    }
  ]
}

{
  "error": "<string>",
  "code": "<string>"
}

{
  "error": "<string>",
  "code": "<string>"
}

documentation

sdk

integrations

cookbook

api reference

create a streaming session

Authorizations

Body

Response