Method, apparatus, and computer program product for reducing the load on a text-to-speech converter in a messaging system capable of text-to-speech conversion of e-mail documents

• US 6,600,814 B1
• Filed: 09/27/1999
• Issued: 07/29/2003
• Est. Priority Date: 09/27/1999
• Status: Expired due to Term

• Current Assignee: Unisys Corporation
• Original Assignee: Unisys Corporation
• Inventors: Shangarpawar, Sachin, Carter, Paul L.
• Primary Examiner(s): Tsang, Fan
• Assistant Examiner(s): Foster, Roland G.
• Application Number:
  US09/406,155
• Time in Patent Office: 1,401 Days
• Field of Search: 379/88.06, 379/88.09, 379/88.13, 379/88.14, 379/88.16, 379/88.17, 379/88.22, 379/88.23, 379/88.25, 379/88.27, 379/88.28, 379/93.01, 379/93.17, 379/93.24, 704/200, 704/258, 704/260, 709/200, 709/204, 709/206
• US Class Current: 379/88.16
• CPC Class Codes: H04M 1/6505 storing speech in digital form
• CPC Class Codes: H04M 2201/60 Medium conversion
• CPC Class Codes: H04M 2203/4536 Voicemail combined with text-based messaging
• CPC Class Codes: H04M 3/5322 for recording text messages

Abstract

A messaging system comprises a storage unit for storing e-mail messages, a text-to-speech converter for converting the different text segments of e-mail messages into speech signals for playback to a user via a telephone handset, and a cache for storing the speech signals of selected ones of previously converted text segments. Upon a subsequent request by a user to convert the text segments of a new e-mail message to speech signals for playback via a telephone handset, the speech signals of previously converted text segments that are identical to any text segments of the new e-mail message are played back from the cache thus avoiding the need for the text-to-speech converter to convert those text segments of the new e-mail message to speech. The load on the text-to-speech converter is thereby reduced.

Claims

1. A messaging system comprising:

an interface that provides access to the system from a telephone handset;

a storage unit for storing e-mail messages, wherein each e-mail message comprises a plurality of different text segments;

a text-to-speech converter for converting the text segments of an e-mail message into speech signals for playback to a user over a telephone handset; and

a cache for storing the speech signals of selected ones of previously converted text segments, wherein upon a request by a user to convert the text segments of a new e-mail message to speech signals for playback via a telephone handset, the speech signals of previously converted text segments that are identical to any text segments of the new e-mail message are played back from the cache thus avoiding the need for the text-to-speech converter to convert those text segments of the new e-mail message to speech and wherein said selected ones of previously converted message body text segments comprise message body text segments having a length that is equal to or lesser than a specified predetermined maximum length.

2. A messaging system comprising:

an interface that provides access to the system from a telephone handset;

a storage unit for storing e-mail messages, wherein each e-mail message comprises a plurality of different text segments;

a text-to-speech converter for converting the text segments of an e-mail message into speech signals for playback to user over a telephone handset; and

a cache for storing the speech signals of selected ones of previously converted text segments, wherein upon a request by a user to convert the text segments of a new e-mail message to speech signals for playback via a telephone handset, the speech signals of previously converted text segments that are identical to any text segments of the new e-mail message are played back from the cache thus avoiding the need for the text-to-speech converter to convert those text segments of the new e-mail message to speech, wherein said selected ones of previously converted message body text segments comprise message body text segments having a length that is equal to or lesser than a predetermined maximum length and wherein said predetermined maximum length is forty (40) characters.

3. A messaging system comprising:

an interface that provides access to the system from a telephone handset;

a storage unit for storing e-mail messages, wherein each e-mail message comprises a plurality of different text segments;

a text-to-speech converter for converting the text segments of an e-mail message into speech signals for playback to user over a telephone handset; and

a cache for storing the speech signals of selected ones of previously converted text segments, wherein upon a request by a user to convert the text segments of a new e-mail message to speech signals for playback via a telephone handset, the speech signals of previously converted text segments that are identical to any text segments of the new e-mail message are played back from the cache thus avoiding the need for the text-to-speech converter to convert those text segments of the new e-mail message to speech,

wherein each e-mail message comprises a TO segment identifying a sender of the e-mail message, a FROM segment identifying a recipient of the e-mail message, a CC segment identifying a person that has been copied on the e-mail message, a RE segment identifying the subject of the e-mail message, and a MESSAGE segment comprising a body of the e-mail message, and wherein said selected ones of previously converted text segments comprises all TO, FROM, CC, and RE segments and wherein each MESSAGE segment comprises one or more sentences, and wherein each sentence of each MESSAGE segment comprises a separate text segment, and further wherein said selected ones of previously converted text segments further comprises each sentence of each MESSAGE segment.

4. The apparatus recited in claim 3, wherein said selected ones of previously converted text segments further comprises any MESSAGE segments having a length that is equal to or lesser than a specified predetermined maximum length.

5. A method for use in a messaging system that comprises a storage unit for storing e-mail messages, each comprising a plurality of different text segments, and a text-to-speech converter for converting the text segments of an e-mail message into speech signals for playback to a user via a telephone handset, said method comprising:

storing the speech signals of selected ones of previously converted text message segments in a cache;

receiving a request from a user to convert the text segments of a new e-mail message to speech signals for playback to the user over a telephone handset;

comparing a text segment of the new e-mail message to the previously converted text segments for which speech signals are stored in the cache; and

if one of said previously converted text segments matches the text segment of the new e-mail message, playing back the stored speech signal for the previously converted text segment from the cache instead of performing a text-to-speech conversion on the text segment of the new e-mail message;

wherein said storing step comprises storing the speech signals of converted text message segments having a length that is equal to or lesser than a specified predetermined maximum length wherein the maximum length applies only to a message body text segment.

7. The method recited in claim 5, wherein each e-mail message comprises a TO segment identifying a sender of the e-mail message, a FROM segment identifying a recipient of the e-mail message, a CC segment identifying a person that has been copied on the e-mail message, a RE segment identifying the subject of the e-mail message, and a MESSAGE segment comprising a body of the e-mail message, and wherein said storing step comprises storing the speech signals of all TO, FROM, CC, and RE segments, wherein each MESSAGE segment comprises one or more sentences, and wherein each sentence of each MESSAGE segment comprises a separate text segment, and further wherein said storing step further comprises storing the speech signals of each sentence of each MESSAGE segment, separately.

8. The method recited in claim 7, further comprising the step of discarding the stored speech signals of the longest sentences as the cache becomes full in order to make space available for caching the speech signals of newer sentences of shorter length.

9. The method recited in claim 8, wherein said discarding step is based on a least recently used algorithm in addition to sentence length.

6. A method for use in a messaging system that comprises a storage unit for storing e-mail messages, each comprising a plurality of different text segments, and a text-to-speech converter for converting the text segments of an e-mail message into speech signals for playback to a user via a telephone handset, said method comprising:

storing the speech signals of selected ones of previously converted text message segments in a cache;

receiving a request from a user to convert the text segments of a new e-mail message to speech signals for playback to the user over a telephone handset;

comparing a text segment of the new e-mail message to the previously converted text segments for which speech signals are stored in the cache; and

if one of said previously converted text segments matches the text segment of the new e-mail message, playing back the stored speech signal for the previously converted text segment from the cache instead of performing a text-to-speech conversion on the text segment of the new e-mail message;

wherein said storing step comprises storing the speech signals of converted text message segments having a length that is equal to or lesser than a predetermined maximum length and wherein said predetermined maximum length is forty (40) characters wherein the maximum length applies only to a message body text segment.

10. A computer-readable medium having program code stored thereon for use in a messaging system that comprises a storage unit for storing e-mail messages, each comprising a plurality of different text segments, and a text-to-speech converter for converting the text segments of an e-mail message into speech signals for playback to a user via a telephone handset, said program code, when executed by a controller, causing the controller to perform the following steps:

store the speech signals of selected ones of previously converted text message segments in a cache;

receive a request from a user to convert the text segments of a new e-mail message to speech signals for playback to the user over a telephone handset;

compare a text segment of the new e-mail message to the previously converted text segments for which speech signals are stored in the cache; and

if one of said previously converted text segments matches the text segment of the new e-mail message, play back the stored speech signal for the previously converted text segment from the cache instead of performing a text-to-speech conversion on the text segment of the new e-mail message, wherein said program code causes only the speech signals of converted text message segments having a length that is equal to or lesser than a specified predetermined maximum length to be stored in the cache wherein the maximum length applies only to a message body text segment.

11. A computer-readable medium having program code stored thereon for use in a messaging system that comprises a storage unit for storing e-mail messages, each comprising a plurality of different text segments, and a text-to-speech converter for converting the text segments of an e-mail message into speech signals for playback to a user via a telephone handset, said program code, when executed by a controller, causing the controller to perform the following steps:

store the speech signals of selected ones of previously converted text message segments in a cache;

receive a request from a user to convert the text segments of a new e-mail message to speech signals for playback to the user over a telephone handset;

compare a text segment of the new e-mail message to the previously converted text segments for which speech signals are stored in the cache; and

if one of said previously converted text segments matches the text segment of the new e-mail message, play back the stored speech signal for the previously converted text segment from the cache instead of performing a text-to-speech conversion on the text segment of the new e-mail message, wherein said program code causes only the speech signals of converted text message segments having a length that is equal to or lesser than a predetermined maximum length to be stored in the cache and wherein said predetermined maximum length is forty (40) characters wherein the maximum length applies only to a message body text segment.

12. A computer-readable medium having program code stored thereon for use in a messaging system that comprises a storage unit for storing e-mail messages, each comprising for plurality of different text segments, and a text-to-speech converter for converting the text segments of an e-mail message into speech signals for playback to a user via a telephone handset, said program code, when executed by a controller, causing the controller to perform the following steps:

store the speech signals of selected ones of previously converted text message segments in a cache;

receive a request from a user to convert the text segments of a new e-mail message to speech signals for playback to the user over a telephone handset;

compare a text segment of the new e-mail message to the previously converted text segments for which speech signals are stored in the cache; and

if one of said previously converted text segments matches the text segment of the new e-mail message, play back the stored speech signal for the previously converted text segment from the cache instead of performing a text-to-speech conversion on the text segment of the new e-mail message, wherein each e-mail message comprises a TO segment identifying a sender of the e-mail message, a FROM segment identifying a recipient of the e-mail message, a CC segment identifying a person that has been copied on the e-mail message, a RE segment identifying the subject of the e-mail message, and a MESSAGE segment comprising a body of the e-mail message, and wherein said program code causes the speech signals of all TO, FROM, CC, and RE segments to be stored in the cache, wherein each MESSAGE segment comprises one or more sentences, and wherein each sentence of each MESSAGE segment comprises a separate text segment, and further wherein said program code causes the speech signals of each sentence of each MESSAGE segment to be stored separately in the cache.

13. The computer-readable medium recited in claim 12, wherein said program code causes the stored speech signals of the longest sentences to be discarded as the cache becomes full in order to make space available for caching the speech signals of newer sentences of shorter length.

14. The computer-readable medium recited in claim 13, wherein said program code causes the stored speech signals to be discarded based upon a combination of sentence length and a least recently used algorithm.