SFSpeechRecognizer 多种语言 [英] SFSpeechRecognizer multiple languages

问题描述

我正在构建一个支持语音识别并将语音转换为文本的搜索，因此我使用了 SFSpeechRecognizer.但是问题是我需要同时支持多种语言比如 ("en_US", "fr", vi, <代码>ar).

I am building a search that supports voice recognition and transforms speech to text so I am using SFSpeechRecognizer. But the problem is that I need to support multiple languages at the same time such as ("en_US", "fr", vi, ar).

主要想法是用户可以用英语说一个单词，另一个用法语说，我希望引擎检测到这一点.

The main idea is that the user can speak for example 1 word in English and the other in French and I want the engine to detect this.

目前，我正在使用它来设置我的主要语言

Currently, I am using this to set my main language

英文:

private let speechRecognizer = SFSpeechRecognizer(locale: Locale.init(identifier: "en_US"))!

法语:

 private let speechRecognizer = SFSpeechRecognizer(locale: Locale.init(identifier: "fr"))!

每种语言我都需要单独设置.

Every language I need to set it separately.

有没有办法让 SFSpeechRecognizer 同时支持多种语言?

Is there a way so SFSpeechRecognizer supports multiple languages at the same time?

推荐答案

老问题，但还没有答案...

Old question, but no answer yet...

您不能在一个 SFSpeechRecognizeer 中使用更多的语言环境，但您可以同时管理多个 SFSpeechRecognizer，如下所示:

You cannot use more locale in just one SFSpeechRecognizeer, but you can manage multiple SFSpeechRecognizers at the same time, like this:

let italian = SFSpeechRecognizer(locale: Locale(identifier: "it-IT"))
let english = SFSpeechRecognizer(locale: Locale(identifier: "en-US"))

为所有这些设置委托:

italian.delegate = self
english.delegate = self

根据需要创建双重、三重或多重逻辑，例如:

create double, triple or multiple logic as needed like:

self.recognitionTaskOne = italian?.recognitionTask(with: recognitionRequest, resultHandler: { (result, error) in
     // Your code
})
self.recognitionTaskTwo = english?.recognitionTask(with: recognitionRequest, resultHandler: { (result, error) in
     // Your code
})

等等...

对于您的具体要求:

主要思想是用户可以用英语说例如1个单词另一个是法语，我希望引擎检测到这一点.

The main idea is that the user can speak for example 1 word in English and the other in French and I want the engine to detect this.

我担心您需要处理所有输出并进行比较.无论如何，我注意到如果我设置了一个语言环境(例如意大利语)然后用户说英语，系统也会得到一些英语.这仅适用于英语...我不确定，但我想无论您设置什么语言环境，默认情况下总是可以理解英语.

I'm affraid that you need to process all the outputs and compare them. Anyway, I noticed that if I set one locale (for example Italian) and then the user speaks in English, the system gets some English too. This works just for English... I'm not sure, but I suppose that whatever locale you set, English is always understood by default.

这篇关于SFSpeechRecognizer 多种语言的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！