简体   繁体   English

SFSpeechRecognizer无法正确处理上下文字符串

[英]SFSpeechRecognizer not handling contextual strings properly

SFSpeechRecognizer is very inaccurate for my use case. 对于我的用例, SFSpeechRecognizer非常不准确。 I have tried setting contextual strings as I have an array of words against which to match with. 我已经尝试设置contextual字符串,因为我有一个与之匹配的单词数组。 But it is still not able to do so. 但它仍然无法做到这一点。 Do I need any other configuration? 我还需要其他配置吗?

I am using sample project provided by Apple with only this change: 我正在使用Apple提供的示例项目,只有这个改变:

recognitionRequest.contextualStrings = @[@"iron man", @"metal", @"stark", @"superhero", @"boyfriend", @"pepper", @"arrogant", @"stylish"];

My understanding is that contextualStrings are only for adding words, that presumably aren't already there, to the dictionary. 我的理解是contextualStrings仅用于在字典中添加可能不存在的单词。

For example : rare/strange words like "Expecto Patronum" or "Fuddle duddle" 例如: rare/strange词,如"Expecto Patronum""Fuddle duddle"

See more in Apple Documentation 请参阅Apple文档中的更多内容

声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.

 
粤ICP备18138465号  © 2020-2024 STACKOOM.COM