Indexing Sound Files on Search Engines that Can’t Hear Them

Référence bibliographique:

SLAWSKI, Bill. Indexing Sound Files on Search Engines that Can’t Hear Them. Creative Flow, May 27th, 2004. URL: http://blog.cre8asite.net/archives/125

Text:

"What do you do if most of the content your company creates is in audio or video format, and you want to include it on the web, and make it so that people can find it? And the content is news, which relies upon timely delivery?

In the case of National Public Radio (NPR), the answer was to put the audio files online, and also offer transcriptions of them. Since NPR started doing that a few weeks ago, they’ve noticed a substantial increase in traffic to their site for topical subjects from the search engines.

One of the things I like about this practice is that people with hearing disabilities are now able to access stories that they couldn’t hear on the radio. What a great result.

The transcription is presently done by speech recognition technology to get stories online very quickly after they are broadcast on the radio. It’s likely that humans will take over from the software presently used, which sometimes garbles results.

If at some point, the search engines become capable of indexing audio, I hope that sites providing transcripts continue to do so. It’s great to see such a great improvement in accessibility, even if it is done inadvertently."

Dublin Core Metadata:

Title: Indexing Sound Files on Search Engines that Can’t Hear Them
Creator: Bill SLAWSKI
Subject: Audio file / audio format / video format / indexing audio / National public radio / recognition / speech recognition.
Description: "What do you do if most of the content your company creates is in audio or video format, and you want to include it on the web, and make it so that people can find it? And the content is news, which relies upon timely delivery?"
Contributor: -
Date: 2004/05/27
Type: Article
Identifier: http://blog.cre8asite.net/archives/125
Source: Creative Flow
Language: En
Coverage: World
Rights: -

An Online Audio Indexing System

Bibliographic reference:

AJMERA, Jitendra. McCOWAN, Iain. BOURLARD, Herve / An Online Audio Indexing System. IDIAP, Switzerland. 2002. URL: ftp://ftp.idiap.ch/pub/reports/2003/rr03-39b.pdf

Text:

Sumary: "This paper presents overview of an online audio indexing system which creates a searchable index of speech content embedded in digitized audio files. This system is based on our recently proposed offline audio segmentation techniques. As the data arrives continuously, the system first finds boundaries of the acoustically homogenous segments. Next, each of these segments is classified as speech, music or \it mixture classes, where mixtures are defined as regions where speech and other non-speech sounds are present simultaneously and noticeably. The speech segments are then clustered together to provide consistent speaker labels. The speech and mixture segments are converted to text via an ASR system. The resulting words are time-stamped together with other metadata information (speaker identity, speech confidence score) in an XML file to rapidly identify and access target segments. In this paper, we analyze the performance at each stage of this audio indexing system and also compare it with the performance of the corresponding offline modules."

URL: ftp://ftp.idiap.ch/pub/reports/2003/rr03-39b.pdf

Dublin Core Metadata:

Title: An Online Audio Indexing System
Creator: AJMERA Jitendra, McCOWAN Iain, BOURLARD Herve
Subject: audio files, audio indexing, automatic speech recognition, speaker clustering.
Description: "This paper presents an overview of an online audio indexing systemwhitch creates a searchable index of speech content embedded in digitized audio files."
Contributor: -
Date: 2004
Type: Paper
Format: Pdf
Identifier:URL: ftp://ftp.idiap.ch/pub/reports/2003/rr03-39b.pdf
Source: IDIAP Research Institute
Language: En
Coverage: World
Rights: IDIAP