|
How long is my free trial? |
You have 30 days to explore the full features of SwiftBridge. |
|
How do I know if my trial has ended? |
After your trial expires, you won’t be able to log in to your SwiftBridge AI account. To reactivate access, please contact our support team. |
|
How do I buy more tokens? |
Request a top up through the platform. 2. Click on Billing and Token Usage. 3. You’ll see your current character balance at the top of the page, along with a button called Request tokens. 4. This will take you to a request form for you to fill in. Please select “Purchase additional tokens / トークンの追加” from the drop-down options. 5. Our team will process your request and top up your tokens. |
|
How many languages does SwiftBridge support? |
Currently: Japanese to English |
|
How long does quality evaluation take? |
Most files with less than 10,000 characters will take around 10 minutes. Larger files will take longer. |
|
How does human verification work? |
Your document is automatically sent for human verification. A Japanese/English financial linguist assesses your AI translation and uses their human expertise to ensure you receive high quality results. |
|
What is sent for human verification? |
All segments undergo human verification. AI improves the initial translation quality, enabling faster vendor processing, thus reducing costs for customers. |
|
Are there any limits on document length or number of files I can submit? |
There is a file limit of 200MB per project. We will be expanding on this more in the near future. One document can be processed at a time. |
|
Why does character count differ between SwiftBridge and other document processors? |
This is due to differences in how tokenizers handle text. For example, Word may split a Japanese character into two tokens while our system counts it as one, or vice versa. Straker’s tokenizer also includes headers and footers in its count. |
| Why can the character count differ between SwiftBridge and other document processors? | This occurs due to differences in how different tokenizers process text. For instance, Word’s tokenizer might treat a particular Japanese character as two separate units, whereas our system may recognize it as a single token—or the other way around. |