我一直在网上冲浪寻找Air for android和ios的离线语音识别解决方案，并且刚刚遇到这个代码。 我想知道如何修改此代码以创建一个实际上是语音识别游戏的应用程序。
我应该定义/添加一个＆＃34;句子＆＃34;到每个帧并检查用户语音输入是否匹配＆＃34;句子＆＃34;或不......
例如，我们将这句话添加到第10帧：＆＃34;这是一个hello world test。＆＃34;如果时间轴到达第10帧，那么：
1）语音识别自动开始。
2）用户的语音转换为文本。
3）如果用户输入匹配＆＃34;这是一个hello world test。＆＃34;然后我们继续框架11
4）如果用户的输入不匹配，则语音识别再次开始（让用户第二次尝试......最后一次尝试）
5）这次如果用户的输入与我们的句子匹配，则继续第11帧，如果不匹配则转到第1帧
我不需要包含在代码或任何控制器中的任何文本到语音，我只想将英语作为我对语音引擎的演讲。非常感谢任何帮助或建议。
这是代码和源文件的链接：
https://fabricemontfort.com/voice-recognition-and-speech-synthesis-in-as3-with-air/
以下是代码：
<?xml version="1.0" encoding="utf-8"?>
<f:Application xmlns:f="library://ns.feathersui.com/mxml"
xmlns:fx="http://ns.adobe.com/mxml/2009"
theme="feathers.themes.MetalWorksMobileTheme">
<fx:Script>
<![CDATA[
import feathers.controls.Alert;
import starling.events.Event;
/**
* STT and TTS AIR Native Extensions
*/
import com.fabricemontfort.air.ezSTT;
import com.fabricemontfort.air.ezSpeech;
import com.fabricemontfort.air.ezspeech.languages;
import com.fabricemontfort.air.ezstt.STTEvent;
import com.fabricemontfort.air.ezstt.languages;
/**
* Text To Speech engine
*/
private var tts:ezSpeech = ezSpeech.instance;
/**
* Speech To Text engine
*/
private var stt:ezSTT = ezSTT.instance;
/**
* actionButton Labels
*/
private static const OUCH:String = "NOT SUPPORTED";
private static const AUTH:String = "ASK PERMISSIONS";
private static const START:String = "START LISTENING";
private static const STOP:String = "STOP LISTENING";
/**
* STT/TTS utterance
*/
private var _utterance:String = "";
[Bindable]
public function get utterance():String {
return _utterance;
}
public function set utterance(value:String):void {
_utterance = value;
resultLabel.text = _utterance;
}
/**
* STT Microphone volume
*/
private var _volume:Number = 0;
[Bindable]
public function get volume():Number {
return _volume;
}
public function set volume(value:Number):void {
_volume = value;
}
/**
* Initialize TTS speed
*/
private var _speed:Number = 0.48;
[Bindable]
public function get speed():Number {
return _speed;
}
public function set speed(value:Number):void {
_speed = value;
tts.setSpeed(speed);
}
/**
* Initialize TTS pitch
*/
private var _pitch:Number = 0.65;
[Bindable]
public function get pitch():Number {
return _pitch;
}
public function set pitch(value:Number):void {
_pitch = value;
tts.setPitch(pitch);
}
override protected function initialize():void {
super.initialize();
// Hide debug messages for ezSTT and ezSpeech
stt.debug = false;
tts.debug = false;
// Set TTS voice speed
tts.setSpeed(speed);
// Set TTS voice pitch
tts.setPitch(pitch);
// Set default language for engines
stt.setLanguage(com.fabricemontfort.air.ezstt.languages.EN);
tts.setLanguage(com.fabricemontfort.air.ezspeech.languages.US);
// Check if STT is supported
if (stt.isSupported()) {
// Check if STT is authorized
if (stt.isAuthorized()) {
// Add STT listeners for final result, partial result, volume, end of speech
stt.addEventListener(STTEvent.PARTIAL, onSTTResult);
stt.addEventListener(STTEvent.FINAL, onSTTResult);
stt.addEventListener(STTEvent.VOL, onSTTVolume);
stt.addEventListener(STTEvent.STOP, onSTTStop);
// Everything is fine, lets start
actionButton.label = START;
// STT is not autorized
} else {
// Initialize STT listener for permissions
stt.addEventListener(STTEvent.AUTH, onAuth);
actionButton.label = AUTH;
}
// STT is not supported
} else {
actionButton.label = OUCH;
var alert:Alert = Alert.show("STT is not supported", "Error", new ListCollection(
[
{label: "OK"}
]));
}
}
/**
* STT engine recognized words
*/
private function onSTTResult(event:STTEvent):void {
// Set utterance with partial and final result
utterance = event.message;
}
/**
* The microphone volume changed
*/
private function onSTTVolume(event:STTEvent):void {
// Set STT microphone volume
volume = parseInt(event.message);
}
/**
* User stopped speaking or clicked the stop button
*/
private function onSTTStop(event:STTEvent):void {
actionButton.label = START;
volume = 0;
// Wait 1 second before repeating last utterance
setTimeout(repeatUtterance, 1000);
}
private function onAuth(event:STTEvent):void {
// Check if STT is authorized
if (stt.isAuthorized()) {
// Remove this listener
stt.removeEventListener(STTEvent.AUTH, onAuth);
// Add STT listeners for final result, partial result, volume, end of speech
stt.addEventListener(STTEvent.PARTIAL, onSTTResult);
stt.addEventListener(STTEvent.FINAL, onSTTResult);
stt.addEventListener(STTEvent.VOL, onSTTVolume);
stt.addEventListener(STTEvent.STOP, onSTTStop);
// Everything is fine, lets start
actionButton.label = START;
// STT is not autorized
} else {
// Show an error message
var alert:Alert = Alert.show("Please give permissions and retry", "Error", new ListCollection(
[
{label: "OK"}
]));
}
}
private function actionButton_triggeredHandler(event:Event):void {
// Check actionButton label
switch (actionButton.label) {
// Ask for permissions
case AUTH: {
stt.askUserAuthorization();
break;
}
// Start STT engine and change label
case START: {
stt.start();
actionButton.label = STOP;
break;
}
// Stop STT engine and change label
case STOP: {
stt.stop();
actionButton.label = START;
break;
}
}
}
/**
* User changed the current language
*/
private function langPickerList_changeHandler(event:Event):void {
// Set current language for STT and TTS engines
stt.setLanguage(langPickerList.selectedItem.code);
tts.setLanguage(langPickerList.selectedItem.codeSpeech);
}
/**
* User asked to enforce a specific language with country code
*/
private function forceLangTextInput_changeHandler(event:Event):void {
// Try to enfore current language for STT and TTS engines (experimental)
stt.forceLanguage(forceLangTextInput.text);
tts.forceLanguage(forceLangTextInput.text);
}
/**
* User asked to repeat the utterance
*/
private function sayButton_triggeredHandler(event:starling.events.Event):void {
repeatUtterance();
}
private function repeatUtterance():void {
// if utterence is not empty
if (utterance != "" && utterance != "[]") {
// and tts is supported
if (tts.isSupported()) {
// Let's talk
tts.say(utterance);
// tts is not supported
} else {
// show a error message
var alert:Alert = Alert.show("TTS is not supported", "Error", new ListCollection(
[
{label: "OK"}
]));
}
}
}
/**
* User changed the speed Slider
*/
private function speedSlider_changeHandler(event:Event):void {
speed = speedSlider.value;
}
/**
* User changed the pitch Slider
*/
private function pitchSlider_changeHandler(event:starling.events.Event):void {
pitch = pitchSlider.value;
}
]]>
</fx:Script>
<!-- This is a vertical layout with padding and gapping of 10 -->
<f:layout>
<f:VerticalLayout gap="10"
padding="10"/>
</f:layout>
<fx:Declarations>
<!-- Fit the target component to the maximum available space -->
<f:VerticalLayoutData id="fitScreen"
percentWidth="100"
percentHeight="100"/>
<!-- Fit the target component to me maximum width -->
<f:VerticalLayoutData id="fitWidth"
percentWidth="100"/>
<!-- List of all supported languages -->
<f:ListCollection id="langListCollection">
<fx:Object label="English"
code="{com.fabricemontfort.air.ezstt.languages.EN}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.US}"/>
<fx:Object label="French"
code="{com.fabricemontfort.air.ezstt.languages.FR}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.FR}"/>
<fx:Object label="German"
code="{com.fabricemontfort.air.ezstt.languages.DE}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.DE}"/>
<fx:Object label="Italian"
code="{com.fabricemontfort.air.ezstt.languages.IT}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.IT}"/>
<fx:Object label="Spanish"
code="{com.fabricemontfort.air.ezstt.languages.ES}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.ES}"/>
<fx:Object label="Chinese"
code="{com.fabricemontfort.air.ezstt.languages.ZH}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.ZH}"/>
<fx:Object label="Japanese"
code="{com.fabricemontfort.air.ezstt.languages.JA}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.JA}"/>
<fx:Object label="Russian"
code="{com.fabricemontfort.air.ezstt.languages.RU}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.RU}"/>
<fx:Object label="Korean"
code="{com.fabricemontfort.air.ezstt.languages.KO}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.KO}"/>
<fx:Object label="Portuguese"
code="{com.fabricemontfort.air.ezstt.languages.PT}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.PT}"/>
<fx:Object label="Czech"
code="{com.fabricemontfort.air.ezstt.languages.CS}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.CS}"/>
<fx:Object label="Dutch"
code="{com.fabricemontfort.air.ezstt.languages.NL}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.NL}"/>
<fx:Object label="Polish"
code="{com.fabricemontfort.air.ezstt.languages.PL}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.PL}"/>
<fx:Object label="Swedish"
code="{com.fabricemontfort.air.ezstt.languages.SV}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.SV}"/>
<fx:Object label="Turkish"
code="{com.fabricemontfort.air.ezstt.languages.TR}"
codeSpeech="{com.fabricemontfort.air.ezspeech.languages.TR}"/>
</f:ListCollection>
</fx:Declarations>
<f:Panel id="resultPanel"
title="RESULTS"
layoutData="{fitScreen}">
<f:layout>
<f:VerticalLayout padding="10"/>
</f:layout>
<!-- This is where user can read the utterance -->
<f:Label id="resultLabel"
wordWrap="true"
text="{utterance}"
layoutData="{fitScreen}"/>
</f:Panel>
<f:Label text="TTS VOICE SPEED"
layoutData="{fitWidth}"/>
<!-- This is where the user can change the TTS voice speed -->
<f:Slider id="speedSlider"
change="speedSlider_changeHandler(event)"
layoutData="{fitWidth}"
minimum="0"
maximum="1"
step="0.05"
value="{speed}"/>
<f:Label text="TTS VOICE PITCH"
layoutData="{fitWidth}"/>
<!-- This is where the user can change the TTS voice pitch -->
<f:Slider id="pitchSlider"
change="pitchSlider_changeHandler(event)"
layoutData="{fitWidth}"
minimum="0.5"
maximum="1.5"
step="0.05"
value="{pitch}"/>
<f:Label text="STT MICROPHONE VOLUME"/>
<!-- This is where the user can see the STT microphone volume -->
<f:ProgressBar id="volumeBar"
minimum="0"
maximum="12"
value="{volume}"
layoutData="{fitWidth}"/>
<!-- This is where the user can pick a language -->
<f:PickerList id="langPickerList"
change="langPickerList_changeHandler(event)"
focusPadding="20"
dataProvider="{langListCollection}"
layoutData="{fitWidth}"/>
<f:Label id="forceLangLabel"
text="TEST A COUNTRY CODE (EXPERIMENTAL)"
layoutData="{fitWidth}"/>
<!-- This is where the user can enforce a language with country code -->
<f:TextInput id="forceLangTextInput"
padding="20"
change="forceLangTextInput_changeHandler(event)"
layoutData="{fitWidth}"/>
<!-- This is where the user have to click to give permissions, start speaking, stop speaking -->
<f:Button id="actionButton"
padding="20"
triggered="actionButton_triggeredHandler(event)"
layoutData="{fitWidth}"/>
<!-- This is where the user have to click to repeat the utterance -->
<f:Button id="sayButton"
padding="20"
label="REPEAT THE LAST UTTERANCE"
triggered="sayButton_triggeredHandler(event)"
layoutData="{fitWidth}"/>
</f:Application>
答案 0 :(得分：1)
首先感谢您对ezSTT和ezSpeech的信心。
如Organis所示。上面提供的源代码面向AS3（和Feathers MXML）。实际上并没有考虑使用Animate CC的内部机制。
但到目前为止还没有丢失任何东西。如果你想使用时间轴，帧，场景的概念，你很可能混合两种方法。
保留纯AS3基础，用于识别和语音合成，以及句子检查和动画路由。 使用MovieClip中的帧来显示播放器的精彩动画，并通过事件在两者之间建立连接。
这将允许你保持动画＆＃34;在保持源代码尽可能简单的同时进行维护。
我希望这会对你有所帮助。用AS3快乐编码。