Google see its browser as the central part of its future computing devices .
With Chrome OS to be released in US,UK ,India and other countries in
around June 2011 , Google Chrome will be used for everthing from listening
to music to creating office documents .
The latest version of Google Chrome enables support for Speech Recognition,
based on HTML5 standards . You can now specify speech as one of your
inputs in your HTML5 form , record what the user speaks and interpret it .
Speech Recognition is one of the widely unexplored fields and with Google
entering this field ,we will surely find many speech based API's in near future.
How Google Chrome Speech Recognition works ?
Even though there aren't much details on the web regarding the working
of Google's Speech Recognition algorithm , it is roughly based on calculating
the MFCC co-efficient of each and every word Google has in its database ,
when you speak a word and pass it to the Google Server to interpret ,it
matches the MFCC of the word spoken by you with its database and returns
you the closest matching word from the database . What all words might
Google have in its database ?? The greatest weapon Google has is its data
that ranges from the Searching trends to Regional Searching data and this is
a one factor that will make it one of the top Speech Recognition Enterprise.
Google makes use of its huge database to give user specific speech to text
conversion based on the regional data it has .
How to use Google Speech Recognition in Google Maps Application ??
- Create an input tag with text as type like this <input type="text" id="address" x-webkit-speech />
- Create an div in which your Google Map will be displayed .
- Create function to geocode your address .
- Create an marker to display the location on the map .
Complete code below is as follows
<html>
<head>
<title>Google Speech Recognition Mashup Google Maps</title>
<style>
#map_canvas {
height: 400;
width : 600
}
</style>
<script type="text/javascript" src="http://maps.google.com/maps/api/js?sensor=false"></script>
<script type="text/javascript">
function initialize() {
geocoder = new google.maps.Geocoder();
var latlng = new google.maps.LatLng(-34.397, 150.644);
var myOptions = {
zoom: 8,
center: latlng,
mapTypeId: google.maps.MapTypeId.ROADMAP
}
map = new google.maps.Map(document.getElementById("map_canvas"), myOptions);
}
function codeAddress() {
var address = document.getElementById("address").value;
geocoder.geocode( { 'address': address}, function(results, status) {
if (status == google.maps.GeocoderStatus.OK) {
map.setCenter(results[0].geometry.location);
var marker = new google.maps.Marker({
map: map,
position: results[0].geometry.location
});
} else {
alert("Geocode was not successful for the following reason: " + status);
}
});
}
</script>
</head>
<body onload="initialize()">
<div id="map_canvas" height="600" width="600" ></div>
<input type="text" id="address" x-webkit-speech />
<input type="button" value="Show Location on Map" onclick="codeAddress();" />
</body>
</html>
This feature can be integrated with number of websites that have a field for
address in their form .Google Speech Recognition works particularly well
with recognizing addresses , however when it comes to recognizing names
Google still has a lot to improve . One improvement Google could certainly
make is embed extra attributes in input tag that will help it narrow down the
search results . For eg in this example it would have been great if it had given
an attrbute like place,email address or phone numbers , which would make
