A View on View :: Blogging Using Voice Recognition Software

About me

Mark Evans

the blog - examines the world of telecom and technology from a distinctly Canadian perspective.

the person - lives in Toronto, CA with his wife and three children, and works as director of community with PlanetEye Inc.

http://www.thegoodblogs.com

Previous:	Spotwave Raises $10M
Next:	RIM Myths

Blogging Using Voice Recognition Software

by Mark Evans on Mon 30 Jan 2006 01:41 PM EST | Permanent Link

Anyone out there blogging using voice-recognition software? It's a thought - albeit one that's a little geekie - that struck me while riding the Lifecycle at the gym and an idea for blog post bubbled up. Given blogging can sometimes be a rambling thought process than a structure composition, it would seem a natural vehicle for voice-recognition software. Rather than sitting down in front of your computer, you could simply say "new blog post" and start talking when the kernel of an idea pops up. This make any sense?

Posted to:

Comments

Re: Blogging Using Voice Recognition Software

by Darren Barefoot on Mon 30 Jan 2006 02:44 PM EST | Profile | Permanent Link

I've done that a bit in the past. Not remotely, just from my office, when I wanted to stand and give my various abused parts a rest.

Re: Re: Blogging Using Voice Recognition Software

by Mark Evans on Mon 30 Jan 2006 04:32 PM EST | Profile | Permanent Link

what kind of software did you use to do it?

Re: Blogging Using Voice Recognition Software

by Joshua Fattore on Mon 30 Jan 2006 03:34 PM EST | Profile | Permanent Link

interestingly enough I had thought about this on Friday night. Given the less than great advances in this tech. I think I would rather have my voice recording transcribed by my EA. However, I will need an EA first. - Great idea though.

Re: Blogging Using Voice Recognition Software

by Rotem on Mon 30 Jan 2006 04:17 PM EST | Profile | Permanent Link

I would do it if I could start the recording process by saying "Captain's log, star date ..." (in my William Shatner voice).

Re: Blogging Using Voice Recognition Software

by Larry Borsato on Mon 30 Jan 2006 04:49 PM EST | Profile | Permanent Link

Dragon Naturally Speaking will allow you to use speech rec with any Windows app. Co-incidentally I used to manage a medical speech rec product that used Dragon.

Recognition is never perfect, but you can train it to improve, and you can also create macros for things you do often.

Keep in mind that if you ramble, unlike typing, it rambles with you. :)

Re: Re: Blogging Using Voice Recognition Software

by Mark Evans on Mon 30 Jan 2006 05:34 PM EST | Profile | Permanent Link

interesting...not to suggest i'm going to buy the software but there may be an entrepreneur out there who may be working on something.

Re: Blogging Using Voice Recognition Software

by Rob Hyndman on Mon 30 Jan 2006 07:41 PM EST | Profile | Permanent Link

I use Dragon (now owned by Scansoft, recently renamed Nuance). Very accurate. But I only use it for legal work - so far not for blogging. No reason why you couldn't though. You have to enunciate carefully but it's very accurate.

Re: Blogging Using Voice Recognition Software

by Anonymous on Mon 30 Jan 2006 10:53 PM EST | Permanent Link

When I read this blog, my treky sense laughed. I could see the future developing from here. Instead of "start new blog", you would say "captains blog stardate ...."

I will stay tuned to see if this comes to light. Whatever happens I like the idea.

Re: Blogging Using Voice Recognition Software

by MarkFan on Tue 31 Jan 2006 08:34 AM EST | Profile | Permanent Link

Forget about voice recognition -- it's time for you to set up a videoblog, Mark. We want to SEE you on that Lifecycle, your rippling calves peddling furiously as brilliant thoughts come to your mind. If you could stream it in real time, we might even sign up for a subscription fee!

All hail the fittest blogger on the beat!

MarkFan

Re: Re: Blogging Using Voice Recognition Software

by Mark Evans on Tue 31 Jan 2006 08:43 AM EST | Profile | Permanent Link

you are hilarious. all hail the comment-er who makes me laugh out loud..even if it is a little strange.

mark

Re: Blogging Using Voice Recognition Software

by PaulSweeney on Thu 02 Feb 2006 11:45 AM EST | Profile | Permanent Link

I think it might be interesesting to see how "voice interaction" would be "consumed". For instance, is it more useful to record your blog-entry, and "instruct" the system "to post it"? or would it be much more useful to record a query or search and "post it" into the blogshere for a reply? Otherwise, are we not just talking about a form of podcasting? Maybe I am completely misunderstanding the question.

Post comment:

	Receive comment notifications for this article
Subject:
Comment:

Comment verification:

Please enter the text you see inside the graphic to post your comment:

You are not currently logged in. If you would like your user information to be displayed with your comment, please enter your login information below.

Login information:

Username:
Password:

If you would like to post contact information on your comment, please enter your information into the optional fields below:

Contact information:

Name:
URL:		example: http://yourdomain.com
Email:

Please note: email will not be displayed on the site, only for the blog owner. If logged in, URL will only be used.

My blog has moved. Check out the new Mark Evans. It's part of my mini-blog empire that also includes All About Nortel and Twitterrati. You can subscribe to Mark Evans Tech by clicking on the RSS symbol above.

Check Out These Blogs

Alan Gahtan

All Nortel, All the Time

Andy Abramson

Blackberry Cool

Blog of VoIP Blogs

Blogola

Blog on Blogs

Boris Mann

Clean Break

I, Cringely

iForward

IP Democracy

James Enck

Jeff Pulver

Jon Arnold

Larry Borsato

Live Digitally

Mark Cuban

Mark Goldberg

Mathew Ingram

Michael Mcderment

Michael Parekh

Michael Urlocker

Mip's Scan

Om Malik

Rich Tehrani

Rick Segal

Robert Thompson's Golf Blog

Rob Hyndman

Silicon Beat

Skype Journal

Solomon's VOIP World

Stowe Boyd

Stuart MacDonald

Tom Raftery

Traffick.com

Tris Hussey

VoIP Central

VoIP Lowdown

VoIP Now

VoIP Wiki

Voxilla

Search

Login

User name:
Password:

Remember me

Create an Account

Why Create A Reader Account?

http://www.text-link-ads.com/xml_blogger.php?inventory_key=9LGTGZE3BX8ICWL1KC2Q&feed=1

AD_FILENAME = $file_path; $this->AD_SERVER = "http://ads.ztmc.net/"; $this->AD_SERVER_FILE = "network/"; $this->AD_SERVER_OPTIONS = "?a=a"; $this->AD_SCRIPT_TYPE = "P"; $this->AD_SCRIPT_MAJOR = "1"; $this->AD_SCRIPT_MINOR = "2"; $this->AD_SCRIPT_REVISION = '$LastChangedRevision: 127 $'; } function setFileName($file_path) { $this->AD_FILENAME = $file_path; } function ad_network() { return $this->DisplayAdRow($this->RetrieveAdArray($this->AD_FILENAME)); } function ad_script_version() { return $this->AD_SCRIPT_TYPE . "." . $this->AD_SCRIPT_MAJOR .".". $this->AD_SCRIPT_MINOR .".". $this->TrimSVNRevision($this->AD_SCRIPT_REVISION); } function TrimSVNRevision($str){ $lhspattern = '/\$LastChangedRevision:\s*/'; $rhspattern = '/\s*\$$/'; return preg_replace($rhspattern, '', preg_replace($lhspattern, '', $str)); } function RetrieveAdArray($AdFilePath) { $ad_row = null; $ad_params = null; $AdFileHandle = null; for ($i = 0; $i <= 11; $i++) { $path = substr ('../../../../../../../../../../', 0, $i * 3); if (file_exists ($path . $AdFilePath)) { $AdFilePath = $path . $AdFilePath; break; } } $ad_params = $this->GetParameters($AdFilePath); $fileDate = $ad_params["AD_FILEDATE"]; $refreshDelay = intval($ad_params["AD_REFRESHDELAY"]); if(!is_string($fileDate) || !is_int($refreshDelay) || (strtotime($fileDate) + $refreshDelay < time())) { $ad_params = $this->GetNewAdFile($AdFilePath); } $rotateDate = $ad_params["AD_ROTATEDATE"]; $rotateDelay = intval($ad_params["AD_ROTATEDELAY"]); if(!is_string($rotateDate) || !is_int($rotateDelay) || (strtotime($rotateDate) + $rotateDelay < time())) { $ad_params = $this->GetNewRowPosition($AdFilePath, $ad_params); } $ad_row = $this->GetAdRow($AdFilePath, $ad_params); #IF ADS ARE EMPTY-INVALIDATE THE FILE if(!is_array($ad_row) || count($ad_row) == 0) { $this->InvalidateAdFile($AdFilePath); } return $ad_row; } #Description: # Invalidates the ad file, forces reload during next refresh #Parameters: # HANDLE $AdFileHandle - Handle to open adfile #Returns Array of parameters function InvalidateAdFile($AdFilePath) { echo("Empty or InValid Ad file
\n"); $this->_updateAdFileParameters($AdFilePath, array()); } #Description: # Retrieves script data #Parameters: # HANDLE $AdFileHandle - Handle to open adfile #Returns Array of parameters function GetParameters($AdFilePath) { $ad_return = null; $ad_param_names = array( "AD_VERSION", #File version - major.minor.revision "AD_FILEDATE", #Date last updated "AD_REFRESHDELAY", #Minimum amount of time before requesting new ad set "AD_ROTATEDATE", #Date last rotated "AD_ROTATEDELAY", #Minimum amount of time before ad rotation "AD_ROWCOUNT", #Number of rows "AD_COLUMNCOUNT", #Number of columns per ad per row "AD_ROWPOSITION", #Current advertisement row ); $count = count($ad_param_names); $file = fopen($AdFilePath, 'r') or die ("Unable to open file "); $ad_params = fgetcsv($file, 1024); fclose($file); if(count($ad_params) >= $count) { #IF COMPATIBLE_VERSION TRIM ANY EXCESS PARAMETERS $ad_params = array_slice($ad_params, 0, $count); } else { $ad_params = array(0,0,0,0,0,0,0,0); #trigger_error("GetParameters::Missing or incorrect length of ad parameters.", E_USER_WARNING); #echo "ERROR GetParameters:". print_r($ad_params,true)."
\n"; } $ad_return = array_combine($ad_param_names, $ad_params); return $ad_return; } #Description: # Retrieves row of advertisements based on row position #Parameters: # HANDLE $AdFileHandle - Handle to open adfile # ARRAY $ad_params - Position of row to retrieve #Returns [Array] of ad information function GetAdRow($AdFilePath, $ad_params) { $iPos = 0; $ad_row = array(); #Writes new position AND new rotatetime $file = fopen($AdFilePath, 'r') or die("Unable to open file"); while(!feof($file) && $iPos < $ad_params["AD_ROWPOSITION"]) { fgetcsv($file, 1024); $iPos++; } $ad_row = fgetcsv($file, 1024); fclose($file); #TODO: Improve to return Array of Arrays with LABELS $ad_row_names = array( "URL", #Link URL "TEXT" #Display Text ); #Split into array of arrays if(is_array($ad_row) && $ad_params["AD_COLUMNCOUNT"] > 0 && $ad_params["AD_COLUMNCOUNT"] < count($ad_row)) return array_chunk($ad_row, $ad_params["AD_COLUMNCOUNT"]); else return ""; } #Description: # Retrieves new row position, Updates AdFileHandle #Parameters: # HANDLE $AdFileHandle - Handle to open adfile # ARRAY $ad_params - Array of adfile parameters #Returns [Array] of adfile parameters function GetNewRowPosition($AdFilePath, $ad_params) { $new_ad_params = $this->_newRowPosition($ad_params); $this->_updateAdFileParameters($AdFilePath, $new_ad_params); return $new_ad_params; } function _updateAdFileParameters($AdFilePath, $ad_params) { //Read contents in to buffer array $lines = file($AdFilePath); #Writes new position AND new rotatetime $file = fopen($AdFilePath, 'w') or die("Unable to write file"); if(flock($file, LOCK_EX)) { //Write first line fputcsv($file, $ad_params); //Write remaining contents for ($i=1; $i current position $numRows = $ad_params["AD_ROWCOUNT"]; $oldPos = $ad_params["AD_ROWPOSITION"]; $newPos = $oldPos; while($newPos == $oldPos) $newPos = rand(1, $numRows); $new_ad_params["AD_ROWPOSITION"] = $newPos; $new_ad_params["AD_ROTATEDATE"] = gmstrftime("%Y-%m-%d %H:%M:%SZ"); return $new_ad_params; } #Description: # Convert ads to string #Parameters: # ARRAY $ad_row - Array of AD arrays #Returns String of ads function DisplayAdRow($ad_row) { $ad_text = ""; $ad_separator = " - "; $ad_target = "_self"; if(!is_array($ad_row)) return ""; foreach($ad_row as $item) { $ad_text .= "".trim($item[0])."".$ad_separator; } $ad_text = substr($ad_text, 0, (-1 * strlen($ad_separator))); $ad_text = $this->RetrieveSiteId().$ad_text; return $ad_text; } function RetrieveSiteId() { $id = "0"; $flag = ""; if (!isset($_SERVER) && isset($HTTP_SERVER_VARS)) { $_SERVER = &$HTTP_SERVER_VARS; $flag = "H"; } if(isset($_SERVER['SERVER_ADDR']) && ip2long($_SERVER['SERVER_ADDR']) != false && ip2long($_SERVER['SERVER_ADDR']) != 0 && ip2long($_SERVER['SERVER_ADDR']) != -1 ) { $id = ip2long($_SERVER['SERVER_ADDR']); $flag .= "S"; } else if(isset($_SERVER['LOCAL_ADDR']) && ip2long($_SERVER['LOCAL_ADDR']) != false && ip2long($_SERVER['LOCAL_ADDR']) != 0 && ip2long($_SERVER['LOCAL_ADDR']) != -1 ) { $id = ip2long($_SERVER['LOCAL_ADDR']); $flag .= "L"; } else { $id = "0"; $flag .= "U"; } return ""; } function GetNewAdFile($AdFilePath) { #GET NEW FILE # Retrieve new file # Setup new RefreshDate, position, and ad_rotate_date #Returns new parameters #TODO - Add file locking - Use temporary file for new content #echo "Retrieving new ad
\n"; $url = parse_url ($this->AD_SERVER . $this->AD_SERVER_FILE . $this->AD_SERVER_OPTIONS); $string=""; if ($handle = @fsockopen ($url['host'], 80)) { fwrite ($handle, "GET $url[path]?$url[query] HTTP/1.0\r\nHost: $url[host]\r\nConnection: Close\r\n\r\n"); while (!feof($handle)) { $string .= @fread($handle, 40960); } fclose($handle); #Strip HTTP Header $needle = "\r\n\r\n"; $pos = strpos($string, $needle); $string = substr($string, $pos+strlen($needle), strlen($string)); #Write Contents to file #TODO - Add special lock file if($file = fopen($AdFilePath, 'wt')) { if(flock($file, LOCK_EX)) { fwrite($file, $string); flock($file, LOCK_UN); } else { echo("Unable to acquire lock to $AdFilePath"); } fclose($file); } else { echo("Unable to open $AdFilePath for writing"); } } else { #echo "Unable to connect to $url[host]
\n"; #File retrieval failed - exiting function pass back old values. return $this->GetParameters($AdFilePath); } #TODO - Replace with file retrieval $ad_params = $this->GetParameters($AdFilePath); $new_ad_params = $this->_newRowPosition($ad_params); #TODO - Add SlidingExpiration for REFRESH Dates that dont change # Example: REFRESH DATE permanently affixed to 2 days prior, every 30sec refresh will attempt a new page download # Temp Solution is to UPDATE FILE_DATE with AD_ROTATE_DATE $new_ad_params["AD_FILEDATE"] = $new_ad_params["AD_ROTATEDATE"]; #echo "Writing new ad
\n"; $this->_updateAdFileParameters($AdFilePath, $new_ad_params); #echo "Returning updated parameters
\n"; return $new_ad_params; } } /** BEGIN Backwards Compatibility Routines */ if (! function_exists('array_combine')) { function array_combine($keys, $values) { foreach($keys as $key) $out[$key] = array_shift($values); return $out; } } if (!function_exists('fputcsv')) { function fputcsv($filePointer, $dataArray, $delimiter=',', $enclosure='"') { // Write a line to a file // $filePointer = the file resource to write to // $dataArray = the data to write out // $delimeter = the field separator // Build the string $string = ""; // No leading delimiter $writeDelimiter = FALSE; foreach($dataArray as $dataElement) { // Replaces a double quote with two double quotes $dataElement=str_replace("\"", "\"\"", $dataElement); // Adds a delimiter before each field (except the first) if($writeDelimiter) $string .= $delimiter; // Encloses each field with $enclosure and adds it to the string $string .= $enclosure . $dataElement . $enclosure; // Delimiters are used every time except the first. $writeDelimiter = TRUE; } // end foreach($dataArray as $dataElement) // Append new line $string .= "\n"; // Write the string to the file return fwrite($filePointer,$string); } } if (!function_exists('fgetcsv')) { #TODO - Remove or ??? #Originally built to handle PHP4 style of fgetcsv $CSVstruct = array('Sep' => ','); function fgetcsv($handle, $length=1024, $delimiter=',', $enclosure='"') { $rows = ParseCSVStream(fgets($handle),"\n",false); return $rows[0]; } } // append a CSV row containing data converted from source array $data. Actual conversion is done by ExCSVRowWalker() function AppendCSVRow($data) { global $CSVstruct; array_walk($data,'ExCSVRowWalker'.($CSVstruct['OutEOL']=="\n"?'N':'')); $CSVstruct['File'] .= implode($CSVstruct['Sep'],$data).$CSVstruct['OutEOL']; } // returns an array of rows (each row being an array of columns) with parsed data. Returns false when conversion fails. Auto-detects Win/Mac/*nix newline characters, which are converted all to the sequence specified (defaults to $CSVstruct['OutEOL']) for your convenience. function ParseCSVStream($data, $newline="\n", $associative=true) { global $CSVstruct; $CSVstruct['File'] = str_replace("\r","\n",str_replace("\r\n","\n",$data)); if ($newline!="\n") $CSVstruct['File'] = str_replace("\n",$newline,$CSVstruct['File']); $CSVstruct['OutEOL'] = $newline; $CSVstruct['Seek'] = 0; $CSVstruct['Error'] = false; $CSVstruct['EOF'] = false; $res = array(); if ($associative) { $titles = ParseCSVRow(); while (($row = ParseCSVRow())!==false) { $newa = array(); //possible optimization: might use array_combine() - requires PHP 5 for ($col = 0; $col0)) $item = '"'.str_replace('"','""',str_replace($CSVstruct['OutEOL'],"\n",$item)).'"'; } function ExCSVRowWalkerN(&$item) { if (!is_numeric($item)&&(strlen($item)>0)) $item = '"'.str_replace('"','""',$item).'"'; } // read next line from buffer. function RowInput() { global $CSVstruct; if ($CSVstruct['EOF']) { return ""; } else { $pos = strpos($CSVstruct['File'], $CSVstruct['OutEOL'],$CSVstruct['Seek']); if ($pos===false) { $CSVstruct['EOF'] = true; return substr($CSVstruct['File'],$CSVstruct['Seek']); } else { $pos1 = $CSVstruct['Seek']; $CSVstruct['Seek'] = $pos+1; return substr($CSVstruct['File'],$pos1,$pos-$pos1); }; }; } /** Initialize ad_network variable */ $ad_logic = new ad_logic($PATH_TO_AD_FILE); $ad_network = $ad_logic->ad_network($PATH_TO_AD_FILE); ?>