Method and system for enabling a user to obtain information from a text-based web site in audio form

• US 6,983,250 B2
• Filed: 10/22/2001
• Issued: 01/03/2006
• Est. Priority Date: 10/25/2000
• Status: Expired due to Fees

• Current Assignee: ONMOBILE LIVE, INC. (Onmobile Global Limited)
• Original Assignee: NMS Communications Corporation (Live Microsystems, Inc.)
• Inventors: Guedalia, David, Guedalia, Jacob
• Primary Examiner(s): Knepper, David D.
• Application Number:
  US10/021,760
• Publication Number:
  US 20020091524A1
• Time in Patent Office: 1,534 Days
• Field of Search: 704/270, 704/270.1, 704/260
• US Class Current: 704/260
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination

Abstract

A method and system for automatic conversion of text to speech including automatically analyzing a text to define at least one vocabulary domain and carrying out a text-to-speech conversion by employing said at least one vocabulary domain.

Claims

1. A method of enabling a user to obtain information from a text-based web site in audio form, comprising:

A. in a first operation to prepare the text-based web site for delivery in audio form;

(i) accessing content of a text-based web site to collect a vocabulary of textual information appearing therein;

(ii) analyzing the collected vocabulary to determine a plurality of limited vocabulary domains into which the textual information of the web site can be grouped, the textual information of each limited vocabulary domain sharing a content-based closeness metric;

(iii) comparing the limited vocabulary domains with existing recorded audio content to determine whether additional audio content is necessary to deliver the web site in audio form, and if so then obtaining such additional audio content; and

(iv) storing formatting configuration information specifying how to deliver the text-based web site in audio format according to the limited vocabulary domains using the existing and additional audio content; and

B. in a second operation performed upon a user'"'"'s request for audio delivery of textual information from the text-based web site;

(i) obtaining the requested textual information from the text-based web site and parsing the textual information into phrases;

(ii) based on the stored formatting configuration information, mapping the parsed phrases to respective ones of the vocabulary domains and providing each parsed phrase to a corresponding limited vocabulary domain server capable of converting the parsed phrase to an audio component;

(iii) receiving audio components from the limited vocabulary domain servers, the audio component resulting from the conversion of the parsed phrases by the limited vocabulary domain servers; and

(iv) generating audio to the user based on the audio components received from the limited vocabulary domain servers.

2. A method according to claim 1, wherein the content-based closeness metric shared by the textual information of each limited vocabulary domain includes sharing one or more selected words.

3. A method according to claim 1, further comprising:

maintaining a cache of the audio components from the limited vocabulary domain servers; and

prior to providing the parsed phrases to the limited vocabulary domain servers, checking whether audio components for the parsed phrases are present in the cache;

and wherein (i) a given parsed phrase is provided to the corresponding limited vocabulary domain server only if the audio component for the given parsed phrase is not present in the cache, and (ii) the audio is generated to the user based on the audio components from the cache if present therein.

4. A method according to claim 1, wherein the text-based web site includes special audio components to be made available to users satisfying a predetermined criteria, and further comprising:

determining whether the user satisfies the predetermined criteria; and

if the user is determined to satisfy the predetermined criteria, then retrieving the special audio components and generating special audio to the user based on the retrieved audio components.

5. A system for enabling a user to obtain information from a text-based web site in audio form, comprising:

A. an analyzer and vocabulary domain definer operative perform a first operation to prepare the text-based web site for delivery in audio form, the first operation including;

(i) accessing content of a text-based web site to collect a vocabulary of textual information appearing therein;

(ii) analyzing the collected vocabulary to determine a plurality of limited vocabulary domains into which the textual information of the web site can be grouped, the textual information of each limited vocabulary domain sharing a content-based closeness metric;

(iii) comparing the limited vocabulary domains with existing recorded audio content to determine whether additional audio content is necessary to deliver the web site in audio form, and if so then obtaining such additional audio content; and

(iv) storing formatting configuration information specifying how to deliver the text-based web site in audio format according to the limited vocabulary domains using the existing and additional audio content; and

B. text-to-speech converter apparatus operative to perform a second operation upon a user'"'"'s request for audio delivery of textual information from the text-based web site, the second operation including;

(i) obtaining the requested textual information from the text-based web site and parse the textual information into phrases;

(ii) based on the stored formatting configuration information, mapping the parsed phrases to respective ones of the vocabulary domains and providing each parsed phrase to a corresponding limited vocabulary domain server capable of converting the parsed phrase to an audio component;

(iii) receiving audio components from the limited vocabulary domain servers, the audio component resulting from the conversion of the parsed phrases by the limited vocabulary domain servers; and

(iv) generating audio to the user based on the audio components received from the limited vocabulary domain servers.

6. A system according to claim 5, wherein the content-based closeness metric shared by the textual information of each limited vocabulary domain includes sharing one or more selected words.

7. A system according to claim 5, wherein the second operation performed by the text-to-speech converter apparatus further includes:

maintaining a cache of the audio components from the limited vocabulary domain servers; and

prior to providing the parsed phrases to the limited vocabulary domain servers, checking whether audio components for the parsed phrases are present in the cache;

and wherein (i) a given parsed phrase is provided to the corresponding limited vocabulary domain server only if the audio component for the given parsed phrase is not present in the cache, and (ii) the audio is generated to the user based on the audio components from the cache if present therein.

8. A system according to claim 5, wherein the text-based web site includes special audio components to be made available to users satisfying a predetermined criteria, and wherein the second operation performed by the text-to-speech converter apparatus further includes:

determining whether the user satisfies the predetermined criteria; and

if the user is determined to satisfy the predetermined criteria, then retrieving the special audio components and generating special audio to the user based on the retrieved audio components.