Backend API
<div class="speech-to-text-demo">
<div class="speech-to-text-container">
<span>Use voice recognition (speech to text)</span>
<dx-speech-to-text
id="speech-to-text"
[class.animation-disabled]="!animation"
[class.custom-button]="displayMode === 'Custom'"
[startText]="startText"
[stopText]="stopText"
[stylingMode]="stylingMode"
[type]="type"
[hint]="hint"
[disabled]="disabled"
[speechRecognitionConfig]="speechRecognitionConfig"
(onStartClick)="onStartClick()"
(onResult)="onResult($event)"
(onEnd)="onEnd()"
></dx-speech-to-text>
<dx-text-area
id="text-area"
[(value)]="textAreaValue"
[width]="360"
[height]="120"
placeholder="Recognized text will appear here..."
[inputAttr]="{ 'aria-label': 'Recognized Text' }"
></dx-text-area>
<dx-button
text="Clear"
[disabled]="textAreaValue === ''"
(onClick)="onClearButtonClick()"
></dx-button>
</div>
<div class="options">
<div class="caption">Options</div>
<div class="option">
<div>Display Mode</div>
<dx-select-box
[(value)]="displayMode"
[dataSource]="displayModes"
[inputAttr]="{ 'aria-label': 'Display Mode' }"
(onValueChanged)="onDisplayModeChanged($event)"
></dx-select-box>
</div>
<div class="option">
<div>Styling Mode</div>
<dx-select-box
[(value)]="stylingMode"
[dataSource]="stylingModes"
displayExpr="displayValue"
valueExpr="value"
[disabled]="displayMode === 'Custom'"
[inputAttr]="{ 'aria-label': 'Styling Mode' }"
></dx-select-box>
</div>
<div class="option">
<div>Type</div>
<dx-select-box
[(value)]="type"
[dataSource]="types"
displayExpr="displayValue"
valueExpr="value"
[disabled]="displayMode === 'Custom'"
[inputAttr]="{ 'aria-label': 'Type' }"
></dx-select-box>
</div>
<div class="switch">
<dx-switch [(value)]="disabled"></dx-switch>
<span>Disabled</span>
</div>
<div class="option-separator"></div>
<div class="option">
<div>Language</div>
<dx-select-box
[(value)]="language"
[dataSource]="languages"
[inputAttr]="{ 'aria-label': 'Language' }"
(onValueChanged)="updateSpeechRecognitionConfig()"
></dx-select-box>
</div>
<div class="switch">
<dx-switch
[(value)]="interimResults"
(onValueChanged)="updateSpeechRecognitionConfig()"
></dx-switch>
<span>Interim Results</span>
</div>
<div class="switch">
<dx-switch
[(value)]="continuous"
(onValueChanged)="updateSpeechRecognitionConfig()"
></dx-switch>
<span>Continuous Recognition</span>
</div>
<div class="option-separator"></div>
<div class="switch">
<dx-switch [(value)]="animation"></dx-switch>
<span>Animation</span>
</div>
</div>
</div>
import { bootstrapApplication } from '@angular/platform-browser';
import { Component, enableProdMode, provideZoneChangeDetection } from '@angular/core';
import { DxSpeechToTextModule } from 'devextreme-angular/ui/speech-to-text';
import { DxTextAreaModule } from 'devextreme-angular/ui/text-area';
import { DxButtonModule } from 'devextreme-angular/ui/button';
import { DxSelectBoxModule } from 'devextreme-angular/ui/select-box';
import { DxSwitchModule } from 'devextreme-angular/ui/switch';
import notify from 'devextreme/ui/notify';
import { AppService } from './app.service';
declare global {
interface Window {
SpeechRecognition?: any;
webkitSpeechRecognition?: any;
}
}
if (!/localhost/.test(document.location.host)) {
enableProdMode();
}
@Component({
selector: 'demo-app',
templateUrl: './app.component.html',
styleUrls: ['./app.component.css'],
imports: [
DxSpeechToTextModule,
DxTextAreaModule,
DxButtonModule,
DxSelectBoxModule,
DxSwitchModule,
],
})
export class AppComponent {
state: 'initial' | 'listening';
startText: string;
stopText: string;
displayMode: string;
displayModes: string[];
stylingMode: string;
stylingModes: { displayValue: string; value: string }[];
type: string;
types: { displayValue: string; value: string }[];
hint: string;
disabled: boolean;
textAreaValue: string;
language: string;
languages: string[];
langMap: { [key: string]: string };
interimResults: boolean;
continuous: boolean;
animation: boolean;
speechRecognitionConfig: Record<string, unknown>;
constructor(private readonly appService: AppService) {
this.state = 'initial';
this.startText = '';
this.stopText = '';
this.displayModes = this.appService.getDisplayModes();
this.displayMode = this.displayModes[0];
this.stylingModes = this.appService.getStylingModes();
this.stylingMode = this.stylingModes[0].value;
this.types = this.appService.getTypes();
this.type = this.types[2].value;
this.hint = 'Start voice recognition';
this.disabled = false;
this.textAreaValue = '';
this.languages = this.appService.getLanguages();
this.language = this.languages[0];
this.langMap = this.appService.getLangMap();
this.interimResults = true;
this.continuous = false;
this.animation = true;
this.updateSpeechRecognitionConfig();
}
onStartClick() {
if (!window.SpeechRecognition && !window.webkitSpeechRecognition) {
notify({
message: 'The browser does not support Web Speech API (SpeechRecognition).',
type: 'error',
displayTime: 7000,
position: 'bottom center',
width: 'auto',
});
return;
}
this.state = 'listening';
this.hint = 'Stop voice recognition';
if (this.displayMode !== 'Custom') {
return;
}
this.type = 'danger';
}
onResult({ event }) {
const { results } = event;
const resultText = Object.values(results)
.map((resultItem) => resultItem[0].transcript.trim())
.join(' ');
this.textAreaValue = resultText;
}
onEnd() {
this.state = 'initial';
this.hint = 'Start voice recognition';
if (this.displayMode !== 'Custom') {
return;
}
this.type = 'default';
}
onClearButtonClick() {
this.textAreaValue = '';
}
onDisplayModeChanged({ value }) {
if (value === 'Text and Icon') {
this.startText = 'Dictate';
this.stopText = 'Stop';
return;
}
this.startText = '';
this.stopText = '';
if (value === 'Custom') {
this.stylingMode = 'contained';
this.type = this.state === 'initial' ? 'default' : 'danger';
}
}
updateSpeechRecognitionConfig() {
this.speechRecognitionConfig = {
lang: this.langMap[this.language],
interimResults: this.interimResults,
continuous: this.continuous,
};
}
}
bootstrapApplication(AppComponent, {
providers: [
provideZoneChangeDetection({ eventCoalescing: true, runCoalescing: true }),
AppService,
],
});
::ng-deep .speech-to-text-demo {
display: flex;
gap: 20px;
height: 640px;
}
::ng-deep .speech-to-text-container {
display: flex;
flex-direction: column;
row-gap: 16px;
flex-grow: 1;
align-items: center;
justify-content: center;
}
::ng-deep #text-area {
margin-top: 16px;
}
::ng-deep .options {
display: flex;
flex-direction: column;
flex-shrink: 0;
width: 300px;
box-sizing: border-box;
padding: 20px;
background-color: rgba(191, 191, 191, 0.15);
gap: 16px;
}
::ng-deep .caption {
font-weight: 500;
font-size: 18px;
}
::ng-deep .option {
display: flex;
flex-direction: column;
row-gap: 4px;
}
::ng-deep .switch {
display: flex;
align-items: center;
column-gap: 8px;
}
::ng-deep .option-separator {
border-bottom: 1px solid var(--dx-color-border);
}
::ng-deep #speech-to-text.animation-disabled {
animation: none;
}
::ng-deep #speech-to-text.custom-button {
border-radius: 2rem;
}
import { Injectable } from '@angular/core';
import { type DxButtonTypes } from 'devextreme-angular/ui/button';
@Injectable({
providedIn: 'root',
})
export class AppService {
displayModes: string[] = [];
stylingModes: { displayValue: string; value: DxButtonTypes.ButtonStyle }[] = [];
types: { displayValue: string; value: DxButtonTypes.ButtonType }[] = [];
languages: string[] = [];
langMap: { [key: string]: string } = {};
constructor() {
this.displayModes = ['Icon Only', 'Text and Icon', 'Custom'];
this.stylingModes = [
{
displayValue: 'Contained',
value: 'contained',
},
{
displayValue: 'Outlined',
value: 'outlined',
},
{
displayValue: 'Text',
value: 'text',
},
];
this.types = [
{
displayValue: 'Normal',
value: 'normal',
},
{
displayValue: 'Success',
value: 'success',
},
{
displayValue: 'Default',
value: 'default',
},
{
displayValue: 'Danger',
value: 'danger',
},
];
this.languages = [
'Auto-detect',
'English',
'Spanish',
'French',
'German',
];
this.langMap = {
'Auto-detect': '',
'English': 'en-US',
'Spanish': 'es-ES',
'French': 'fr-FR',
'German': 'de-DE',
};
}
getDisplayModes() {
return this.displayModes;
}
getStylingModes() {
return this.stylingModes;
}
getTypes() {
return this.types;
}
getLanguages() {
return this.languages;
}
getLangMap() {
return this.langMap;
}
}
<!DOCTYPE html>
<html lang="en">
<head>
<title>DevExtreme Demo</title>
<meta http-equiv="X-UA-Compatible" content="IE=edge" />
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=5.0" />
<link rel="stylesheet" type="text/css" href="https://cdn3.devexpress.com/jslib/26.1.5/css/dx.light.css" />
</head>
<body class="dx-viewport">
<div class="demo-container">
<demo-app>Loading...</demo-app>
</div>
<script src="https://cdn.jsdelivr.net/npm/zone.js@0.16.2/bundles/zone.umd.js"></script>
<script src="./bundle.js" type="module"></script>
</body>
</html>
You can integrate SpeechToText with any text input, including editors available in the DevExtreme product line. Simply set a component's value property to transcribed text. SpeechToText returns transcribed text in the onResult handler as a user speaks. When speech stops, the component calls the onEnd handler and switches from a "listening" state to the initial state. SpeechToText implements state-dependent icons (startIcon/stopIcon), text (startText/stopText), and click handlers (onStartClick/onStopClick).
For a complete overview of SpeechToText options (including Web Speech API options), refer to the following topic: SpeechToText API Reference.