Translator and Language Detector APIs
The inputQuota read-only property of the LanguageDetector interface returns the input quota available to the browser for detecting languages.
Value
A number specifying the available input quota.
This number is implementation-dependent. For example, it might be Infinity if there are no limits beyond the user's memory and the maximum length of JavaScript strings, or it might be a number of tokens in the case of AI models that use a token/credits scheme.
The only guarantee is that inputQuota - LanguageDetector.measureInputUsage will be non-negative if there is sufficient quota to detect the text's language.
Examples
Checking if you have enough quota
In the below snippet, we create a new LanguageDetector instance using LanguageDetector.create_static, then return the total input quota via inputQuota and the input quota usage for detecting a particular text string's language via LanguageDetector.measureInputUsage.
We then test to see if the individual input usage for that string is greater than the total available quota. If so, we throw an appropriate error; it not, we commence detecting the string's language using LanguageDetector.detect.
const detector = await LanguageDetector.create({
expectedInputLanguages: ["en-US", "zh"],
});
const totalInputQuota = detector.inputQuota;
const inputUsage = await detector.measureInputUsage(myTextString);
if (inputUsage > totalInputQuota) {
throw new Error("Insufficient quota to detect languages.");
} else {
console.log("Quota available to detect languages.");
const results = await detector.detect(myTextString);
// ...
}
Specifications
Browser compatibility
See also
来源:MDN Web Docs / Mozilla 贡献者 · 作者与贡献记录 · CC BY-SA 2.5。离线整理了版式、链接和媒体显示。
