Set text prompt language

Try image generation (Vertex AI Studio)

Try Imagen in a Colab

This page describes how you can set an optional Imagen on Vertex AI parameter to specify the prompt language that you use. If you don't specify a language, then Imagen automatically detects the language.

a book image generated from a prompt in hindi
Image generated from prompt: ऊपर से देखा गया किताबों का ढेर। सबसे ऊपरी पुस्तक में एक पक्षी का जलरंग चित्रण है। किताब पर VERTEX AI मोटे अक्षरों में लिखा हुआ है 1

1 A pile of books seen from above. The topmost book contains a watercolor illustration of a bird. VERTEX AI is written in bold letters on the book.
an image of a woman from a prompt in korean
Image generated from prompt: 어두운 노란색과 청록색으로 이루어진 밝은 색의 옷을입고 귀걸이를 끼고있는 여자 포스트 모던 패션 사진 2

2 Woman wearing bright colors, in the style of dark yellow and dark cyan, wearing earrings, postmodern fashion photography.

Set text prompt language

The following input values are supported for the text-prompt lanague:

  • Chinese (simplified) (zh/zh-CN)
  • Chinese (traditional) (zh-TW)
  • English (en, default value)
  • French (fr)
  • German (de)
  • Hindi (hi)
  • Japanese (ja)
  • Korean (ko)
  • Portuguese (pt)

  • Spanish (es)

Console

If your prompt is in one of the supported languages, Imagen detects and translates your text and returns your generated or edited images.

If your prompt is in an unsupported language, Imagen uses the text verbatim for the request. This might result in unexpected output.

REST

Before using any of the request data, make the following replacements:

  • PROJECT_ID: Your Google Cloud project ID.
  • TEXT_PROMPT: The text prompt that guides what images the model generates. This field is required for both generation and editing.
  • PROMPT_LANGUAGE: string. Optional. The language code that corresponds to your text prompt language. In this example it would be hi. Available values:
    • auto - Automatic detection. If Imagen detects a supported language, the prompt (and optionally, a negative prompt), are translated to English. If the language detected is not supported, Imagen uses the input text verbatim, which might result in unexpected output. No error code is returned.
    • en - English (default value if omitted)
    • es - Spanish
    • hi - Hindi
    • ja - Japanese
    • ko - Korean
    • pt - Portuguese
    • zh-TW - Chinese (traditional)
    • zh or zh-CN - Chinese (simplified)

HTTP method and URL: 

POST https://us-central1-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/publishers/google/models/imagegeneration@005:predict

Request JSON body: 

{
  "instances": [
    {
      "prompt": "सूर्यास्त के समय एक समुद्र तट। उड़ते पक्षी, हवा में लहराते नारियल के पेड़। लोग समुद्र तट पर सैर का आनंद ले रहे हैं।"
    }
  ],
  "parameters": {
    "language": "PROMPT_LANGUAGE"
  }
}

To send your request, choose one of these options:

curl

Save the request body in a file named request.json, and execute the following command: 

curl -X POST \
     -H "Authorization: Bearer $(gcloud auth print-access-token)" \
     -H "Content-Type: application/json; charset=utf-8" \
     -d @request.json \
     "https://us-central1-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/publishers/google/models/imagegeneration@005:predict"

PowerShell

Save the request body in a file named request.json, and execute the following command: 

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred" }

Invoke-WebRequest `
    -Method POST `
    -Headers $headers `
    -ContentType: "application/json; charset=utf-8" `
    -InFile request.json `
    -Uri "https://us-central1-aiplatform.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/publishers/google/models/imagegeneration@005:predict" | Select-Object -Expand Content
The following sample response is for a request with "sampleCount": 2. The response returns two prediction objects, with the generated image bytes base64-encoded. 
{
  "predictions": [
    {
      "bytesBase64Encoded": "BASE64_IMG_BYTES",
      "mimeType": "image/png"
    },
    {
      "mimeType": "image/png",
      "bytesBase64Encoded": "BASE64_IMG_BYTES"
    }
  ]
}

What's next