Cortex Translate
Editions
Production use of this feature is available for specific editions only. Contact our sales team for more information.
The Cortex Translate transformation component uses Snowflake Cortex to translate the input text of one or more columns from one supported language to another—for example, from English to German. When a column is included, all rows in that column will be translated.
For a list of supported languages, read the Usage notes.
To use this component, you must use a Snowflake role that has been granted the SNOWFLAKE.CORTEX_USER database role. Read Required Privileges to learn more about granting this privilege.
To learn more about Snowflake Cortex, such as availability, usage quotas, managing costs, and more, read Large Language Model (LLM) Functions (Snowflake Cortex).
Use case
This component lets you translate content between two languages directly in your pipeline. For example, you can use it to:
- Translate support tickets into the language your customer support teams speak so that they can respond to requests with better efficiency.
- Translate product descriptions for products to be sold online in different countries or regions.
Properties
Name = string
A human-readable name for the component.
Columns = column editor
- Input Column: Select a column from your input table. The text in each row of this column will be translated.
- Source Language: Select the source language that the input text begins as.
- Target Language: Select the target language to translate the input text to.
To translate more than one input column, click +.
Include Input Columns = boolean
- Yes: Includes both your source language input columns and the new target language translation columns. This will also include those input columns not selected in Columns.
- No: Only includes the new target language translation columns.
Example
A company has hired a support engineer, Klaus, whose first language is German. Klaus's English is passable, but he can respond to support queries faster if he can read the queries in German. Klaus's latest support cases look like this:
Input data
_______________________________________________________________________
| DATE | SUPPORT CASE | NAME |
|------------|----------------------------------------------|-----------|
| 2023-01-10 | Unable to access data sources. | Alice |
| 2023-02-05 | Need help with query optimization. | Bob |
| 2023-03-20 | Issue with database schema update. | Charlie |
| 2023-04-15 | How to automate data ingestion? | Dana |
| 2023-05-01 | Data validation failing on new records. | Eve |
| 2023-06-25 | Need guidance on ETL tool selection. | Frank |
| 2023-07-10 | Client reporting login issues. | Grace |
| 2023-08-05 | Data pipeline running slow. | Henry |
| 2023-09-20 | How to handle large data volumes? | Iris |
| 2023-10-15 | Error connecting to external API. | Jack |
| 2023-11-30 | Need assistance with ETL scheduling. | Kate |
| 2023-12-25 | Best practices for data integration. | Liam |
| 2024-01-10 | How to handle real-time data streams? | Mia |
| 2024-02-05 | Data quality issues in reports. | Noah |
| 2024-03-20 | Performance degradation in ETL jobs. | Olivia |
| 2024-04-15 | How to implement data masking? | Peter |
|____________|______________________________________________|___________|
Klaus can use the Cortex Translate component to translate the SUPPORT CASE column from English to German, making it easier for him to understand and respond to support cases.
Translate chosen columns:
Cortex Translate component properties:
- Columns
- Input Column: SUPPORT CASE
- Source Language: English
- Target Language: German
- Include Input Columns: YES
By setting Include Input Columns to YES, the original columns from the table will be kept as part of the pipeline run, and the target language column (in this case, de_SUPPORT CASE) is appended to the end of the table.
Output data
________________________________________________________________________________________________________________________________
| DATE | SUPPORT CASE | NAME |de_SUPPORT CASE |
|------------|----------------------------------------------|-----------|--------------------------------------------------------|
| 2023-01-10 | Unable to access data sources. | Alice |Ich kann auf Datenquellen nicht zugreifen. |
| 2023-02-05 | Need help with query optimization. | Bob |Ich benötige Hilfe bei der Abfragezeichnung. |
| 2023-03-20 | Issue with database schema update. | Charlie |Problem bei der Aktualisierung des Datenbankschemas. |
| 2023-04-15 | How to automate data ingestion? | Dana |Wie automatisiert man die Datenerfassung? |
| 2023-05-01 | Data validation failing on new records. | Eve |Datenvalidierung schlägt bei neuen Datensätzen fehl. |
| 2023-06-25 | Need guidance on ETL tool selection. | Frank |Ich benötige Anleitung zur Auswahl des ETL-Tools. |
| 2023-07-10 | Client reporting login issues. | Grace |Probleme mit der Kundenmeldung. |
| 2023-08-05 | Data pipeline running slow. | Henry |Die Datenpipeline läuft langsam. |
| 2023-09-20 | How to handle large data volumes? | Iris |Wie man mit großen Datenmengen umgeht? |
| 2023-10-15 | Error connecting to external API. | Jack |Fehler bei der Verbindung mit der externen API. |
| 2023-11-30 | Need assistance with ETL scheduling. | Kate |Ich benötige Hilfe bei der ETL-Planung. |
| 2023-12-25 | Best practices for data integration. | Liam |Bewährte Verfahren für die Datenintegration. |
| 2024-01-10 | How to handle real-time data streams? | Mia |Wie verarbeitet man Echtzeit-Datenströme? |
| 2024-02-05 | Data quality issues in reports. | Noah |Datenqualitätsprobleme in Berichten. |
| 2024-03-20 | Performance degradation in ETL jobs. | Olivia |Leistungsabbau bei ETL-Jobs. |
| 2024-04-15 | How to implement data masking? | Peter |Wie implementiert man die Datenmaskierung? |
|____________|______________________________________________|___________|________________________________________________________|