Create Transcription Translation | SoferAi

from soferai import SoferAI
import uuid
client = SoferAI(
    api_key="YOUR_TOKEN_HERE",
)
client.transcribe.create_transcription_translation(
    transcription_id=uuid.UUID("d5e9c84f-c2b2-4bf4-b4b0-7ffd7a9ffc32"),
)

{
  "transcription_id": "d5e9c84f-c2b2-4bf4-b4b0-7ffd7a9ffc32",
  "target_mode": "paragraph",
  "translation_style": "formal",
  "status": "in_progress",
  "completed_segments": 1,
  "failed_segments": 0,
  "segments": [
    {
      "segment_id": "seg-001",
      "segment_index": 1,
      "source_hash": "a1b2c3d4e5f6",
      "status": "completed",
      "is_stale": false,
      "source_language": "en",
      "target_language": "es",
      "start_time": 0,
      "end_time": 15.5,
      "speaker": "Speaker 1",
      "content": "Hello, everyone. Welcome to our presentation.",
      "error": ""
    },
    {
      "segment_id": "seg-002",
      "segment_index": 2,
      "source_hash": "f6e5d4c3b2a1",
      "status": "in_progress",
      "is_stale": false,
      "source_language": "en",
      "target_language": "es",
      "start_time": 15.6,
      "end_time": 30,
      "speaker": "Speaker 2",
      "content": "Thank you for joining us today.",
      "error": ""
    }
  ],
  "model": "whisper-1",
  "total_segments": 2,
  "created_at": "2024-01-15T09:30:00Z",
  "updated_at": "2024-01-15T10:00:00Z",
  "error": ""
}

Start a background paragraph-level translation for a completed transcription

Authentication

AuthorizationBearer

Bearer authentication of the form Bearer <token>, where token is your auth token.

Path parameters

transcription_idUUIDRequired

ID of the transcription to translate.

Request

This endpoint expects an object.

target_modestringOptional

Translation target. Use auto to translate Hebrew/Yiddish/Aramaic-heavy segments to English and English-heavy segments to Hebrew.

translation_stylestringOptional

Translation style id. Built-ins include learning_english and sefer_hebrew; saved user guides use custom:<style_id>.

forcebooleanOptional

Regenerate the translation even if cached segments already exist.

Response

This endpoint returns an object.

transcription_idUUID

ID of the source transcription.

target_modestring

Translation target mode.

translation_stylestring

Translation style id.

statusstring

Overall translation status.

completed_segmentsinteger

Number of successfully translated segments.

failed_segmentsinteger

Number of failed segments.

segmentslist of objects

Paragraph-level translated segments.

modelstringOptional

Model used for translation.

total_segmentsintegerOptional

Total paragraph segments in the translation job.

created_atdatetimeOptional

When the translation job was created.

updated_atdatetimeOptional

When the translation job was last updated.

errorstringOptional

Overall job error, if any.

Errors

401

Authentication Error

404

Transcription Not Found

429

Rate Limit Error