Skip to Content

WebSocket API V1 Configuration

Legacy — V1 remains supported but is not recommended for new integrations. Use V2 for new production work, or evaluate V3 (beta) if you need per-message MP4 delivery or a structured session lifecycle.

Overview

The V1 WebSocket API provides action-based messaging for text-to-sign-language video translation. This is the legacy protocol but remains fully supported.

Connection Details

Connection URL

WebSocket URL
wss://ai.api.production.signapsesolutions.com

Authentication

  • Production: Use platform authentication (API Gateway tokens, cookies, etc.)

Message Schema

Client → Server Request

Client Request
{
"action": "LiveTranslation",
"sentence": "string",
"chat": {
  "uid": "string"
},
"metadata": {
  "responseFormat": "hls"
}
}

Server → Client Response

Server Response
{
"action": "LiveTranslation",
"success": true,
"message": "string"
}

Configuration Options

Required Fields

NameTypeRequiredDescription
actionstringrequired
The action to perform.
TranslateSentenceLiveTranslation
sentencestringrequired
The text sentence to translate into sign language video.

Optional Fields

NameTypeRequiredDescription
metadataobjectoptional
Additional configuration options.
Default: {}
responseFormatstringoptional
Output video format.
Default: hls
hlsmp4
responseDownloadTypestringoptional
Download method (internally set to "presignedUrl").
Default: Auto-set
typestringoptional
Request type based on action: "chat" for TranslateSentence, "live" for LiveTranslation.
Default: Auto-set
connectionstringoptional
Connection type (internally set to "websocket").
Default: Auto-set

Additional custom metadata fields are also passed through to the video generation service.

Actions

LiveTranslation

Streamlined translation for real-time use cases.

Features:

  • No profanity filtering
  • No rate limiting
  • No word count validation
  • Optimized for speed and responsiveness
LiveTranslation Example
{
"action": "LiveTranslation",
"sentence": "Live translation message",
"metadata": {
  "responseFormat": "hls"
}
}

WebSocket Events

The server sends informational messages during processing:

  • Progress updates
  • Moderation notices
  • Processing status
  • Error notifications

These are sent as text messages to the WebSocket connection.

Media Delivery

Video output is delivered via presigned URLs. The WebSocket connection receives:

  • Status acknowledgment messages
  • Processing progress updates
  • Completion notifications

The actual video media is retrieved through presigned URLs generated by backend services according to the specified responseFormat.

HLS Format

For HLS format, clients should:

  1. Receive the presigned URL for the manifest
  2. Poll or watch the HLS manifest/segments
  3. Stream the video content

MP4 Format

For MP4 format, clients should:

  1. Receive the presigned URL for the complete video
  2. Download the video file directly

Migration to V2

Consider migrating to the V2 protocol for:

  • More structured message format
  • Better protocol versioning support
  • Unified content and output configuration
  • Enhanced extensibility

See the V2 Protocol documentation for details.

Last updated on
Question? Give us feedback
support@signapse.ai