GB2405066A

GB2405066A - Auditory assistance with language learning and pronunciation via a text to speech translation in a mobile communications device

Info

Publication number: GB2405066A
Application number: GB0310949A
Authority: GB
Inventors: Jeffrey Wilson
Original assignee: Intellprop Ltd
Current assignee: Intellprop Ltd
Priority date: 2003-05-13
Filing date: 2003-05-13
Publication date: 2005-02-16
Also published as: GB0310949D0

Abstract

A telecommunications services apparatus, suitable for text to speech translation, includes a text message processor 14 for interpreting text content, from a text message input by a user's terminal 10, into textual phrases and/or words. Audio playback equipment 17, such as interactive voice response equipment, in association with a controller 16 and an audio store 18, can then convert the textual phrases and/or words into audio to be played to the user or to a third party in a subsequent telephone voice communication. The invention can be applied to auditory assistance with language learning and pronunciation.

Description

1 2405066

TELECOMMUNICATIONS SERVICES APPARATUS

The present invention concerns the fields of telecommunications, text and voice services and systems for language learning and entertainment.

So called 'Text to Speech' techniques have been used in a number of fields for converting textual input provided by one party into spoken audio for another party.

Examples include voice-mail read back systems and computer accessibility products for the disabled that read out file contents.

The prior art systems are used to convert either stored text, such as in a computer file, or text provided by a third party into spoken audio.

Text messaging systems are known in telecommunications, particularly in the field of mobile telephony, where systems such as GSM support text messaging between handsets in a defined and simple way. Text messaging to network applications is also known, and some text applications are capable of interacting with a user via a text message exchange. A. e À -.e

Interactive voice recognition (IVR) systems are also known where spoken audio prompts are produced by the system, and user input, usually by DTMF but also via dial À.. pulses, spoken words or other sounds, is recognised by the system and used to control the flow of the interaction. Such systems are commonly used as menu driven auto À attendant on the customer service telephone lines of companies. ....

It is also known that auditory assistance with language learning and pronunciation is available in a variety of forms including pre-recorded disks or tapes, television programs and person-to-person teaching.

When learning a language, it is desirable to hear the correct pronunciation of words and phrases. To an extent pronunciation may be derived from the phonetic notation commonly found in some dictionaries, but this is inconvenient and prone to user error or misinterpretation.

It is not possible to use a text-only system to provide improved pronunciation guidance, as it would suffer from the same problems as the dictionary. An automated voice-only system is also impractical, since even with a sophisticated voice recogniser, mispronunciation of input would be likely to result in misrecognition.

Pre-recorded material is inconvenient for finding the pronunciation of a particular word or phrase, since the desired word must first be found amongst the available material.

According to the invention there is provided a method whereby a user inputs text, and the text content is interpreted into textual phrases and/or words which, after being looked up in a database, are converted to audio to be played to the same user in a subsequent voice communication. re Àe

According to a further aspect of the inventive method, the text may be input by sending a text message to the apparatus using a telephone network. À .

According to a further aspect of the invention, the database may provide identifiers of À e.

À audio files containing the audio to be played. À: '

-

According to another aspect of the invention, the database may provide identifiers to enable synthesis of the audio to be played.

According to the invention there is provided a telecommunications services apparatus operable to receive text input from a user, wherein the text content is interpreted into textual phrases and/or words which, after being looked up in a database, are converted to audio to be played to the same user in a subsequent voice communication.

According to a further aspect of the invention, there is provided text message processing apparatus in a mobile telecommunications network operable to receive a text message from a user, and from the text input to derive identifiers of audio to be played. s

According to a further aspect of the invention, the apparatus comprises an audio file store and playback facility operable to play audio files according to identifiers provided by the database.

According to another aspect of the invention, the apparatus comprises an audio synthesizer operable to synthesise speech from the identifiers provided by the database.

Referring to Figure 1, a user (10) sends a text message via an MSC (12). The message is intercepted by a text message processor (14), which interprets the message and passes it to the Controller (16). Subsequently, the user makes a voice call to an Audio playback equipment (17), which determines from the Controller (17) which audio files e. ee : ' . to play to this user from the Audio store (18). À - . À À--e

The present invention provides the ability for a user to send a text message containing À a word or phrase for which he desires to hear the corresponding spoken audio, possibly À.

for assistance with language learning, for entertainment, or for other purposes. The À audio is played to a listener, who is preferably the same user, but may be a third party, during a subsequent telephone call, which may preferably be dialled by the recipient or alternatively outdialled by the equipment. The original text message and the subsequent call are preferably associated with the same user by means of the user's CLI.

In a preferred embodiment, the text input is entered using a text message, for example using the SMS facility of a GSM mobile telephone network. The message is then sent by normal means. The telephone network is arranged, as is increasingly common, so that mobile originated text messages pass through an SMS processing node, usually called an SMS Router. In addition to providing simple message routing functions, the SMS Router is able to examine message content and addressing to provide intelligent routing functions. For this application the message is identified as being in a 'text to speech' category either by means of its address, which may be a short code or other address identifying this service, or by means of the message syntax whereby the content for example begins with a special character such as '&', or by other means.

Once identified as being in the 'text to speech' category, the message is preferably routed away from the SS7 telephone network to a Controller or server. This routing may preferably be done over TCP/IP. The Controller may then store the message, or information extracted from the message, along with the CLI of the user in a database.

The information extracted preferably comprises identifiers of audio to be played to the user in a subsequent telephone call. The identifiers may be file identifiers relating to an audio store, or identifiers permitting synthesis of the required audio, or the like.

Subsequently, the user dials into the service with a voice call. Preferably the telephone number dialled is the same as the destination address used for the original text À- : ' .. message, which may be a short code. Alternatively, the system may dial out to the user, although this has more complex billing aspects.

The voice call is preferably routed by the network to a voice services equipment (VSE) À:e or other apparatus capable of audio playback from files or using synthesis. In a preferred embodiment, an Interactive Voice Response (IVR) equipment may be used, À , . . . which is commonly known in the art.

The VSE communicates with the Controller, or alternatively the Controller may be part of the VSE. The Controller determines the identity of the caller from the CLI of the voice call, and looks up in its database to find whether information has been stored for this user from a previous text message. If so, the stored identifiers are used to play the desired audio to the caller. If there is no CLI match in the database, then an error message may be played to the caller.

Alternatively, the system may be used for a third party to hear the audio specified textually by the user. This could be done in a number of ways. For example, the message syntax could be extended to include the mobile address of a recipient, to which an SMS invitation could be sent. This would invite the recipient to dial a particular number to hear the audio defined in the text message.

The syntax could also be extended to allow specification of the voice type for playback, since for entertainment applications a number of different voice types would be an advantage.

The syntax could be further extended to permit textual identifiers for sounds other than words to be included. These could include a range of sound effects or the audio equivalents of textual 'smileys' which can help to add emotional content to a message.

An example message could be: 0307123456.female2. You passed your exam! *#31# Well Done.

À Sent to short code 123, this message would be identified as a text to speech message . À . by the short code address. The telephone number and '.' separator indicates that a text ^. 20 invitation message is to be sent to this number, inviting the recipient to call 123 to receive a message. When the recipient makes a voice call to 123, they will hear audio played in a particular female voice, specified by 'female2' and a further '.' separator, saying "You passed your exam <Sound of cheering> Well Done", where the cheering is an audio sound effect specified by *#31#.

Many other forms of syntax are possible.

In so far as the embodiment(s) of the invention described above may be implemented, at least in part, using software controlled processing apparatus, it will be appreciated that a computer program providing such software control and a storage medium by which such a computer program is stored are envisaged as aspects of the invention.

Claims

1. Telecommunications services apparatus operable to receive text input from a user, the apparatus comprising means for interpreting the text content into textual phrases and/or words, and means associated with a database for converting the textual phrases and/or words into audio to be played in a subsequent voice communication.

2. Apparatus according to claim 1, including text message processing apparatus operable to derive identifiers of the audio to be played from the received text input.

3. Apparatus according to claim 1 or claim 2, comprising an audio file store and playback means operable to play the audio from audio files according to identifiers provided by the database.

4. Apparatus according to claim 1 or claim 2, comprising an audio synthesiser operable to synthesise speech from identifiers provided by the database.

5. Apparatus according to any one of claims 1 to 4, comprising routing means operable to recognise text input intended for text to speech conversion.

6. Apparatus according to claim 5, wherein the recognition by the routing means is based on the address of the text input.

7. Apparatus according to claim 6, wherein the text input address for text to speech conversion is a short code address.

8. Apparatus according to claim 5, wherein the recognition by the routing means is based on the message syntax of the text input.

9. Apparatus according to claim 8, wherein the message syntax for text to speech conversion is when the message begins with at least one specific character.

10. Apparatus according to any one of claims 1 to 9, wherein the database stores the message in the text input, or information extracted from the message, together with the calling line identity (CLI) of the user.

11. Text message processing apparatus in a mobile telecommunications network operable to receive a text message from a user, and from the text input to derive identifiers of audio to be played.

12. Apparatus substantially as herein described with reference to and as illustrated in the accompanying drawing.

13. A method of converting text into audio, whereby a user inputs text, and the text content is interpreted into textual phrases and /or words which, after being looked up in a database, are converted to audio to be played in a subsequent voice 1 5 communication.

14. A method according to claim 13, wherein the text is input by sending a text message using a telephone network.

15. A method according to claim 13 or claim 14, including text message processing operable to derive identifiers of the audio to be played from the received text input.

16. A method according to claim 13, claim 14 or claim 15, including playing the audio from audio files according to identifiers provided by the database.

17. A method according to claim 13, claim 14 or claim 15, including synthesising speech with an audio synthesiser from identifiers provided by the database.

18. A method according to any one of claims 13 to 17, comprising routing text input recognised as being intended for text to speech conversion.

19. A method according to claim 18, wherein the recognition is based on the address of the text input.

20. A method according to claim 19, wherein the text input address for text to speech conversion is a short code address.

21. A method according to claim 20, wherein the recognition is based on the message syntax of the text input.

22. A method according to claim 21, wherein the message syntax for text to speech conversion is when the message begins with at least one specific character.

23. A method according to any one of claims 13 to 22, wherein the database stores the message in the text input, or information extracted from the message, together with the calling line identity (CLI) of the user.

24. A method of text message processing in a mobile telecommunications network operable to receive a text message from a user, and from the text input to derive identifiers of audio to be played.

25. A method substantially as herein described with reference to and as illustrated in the accompanying drawing.

26. A computer program for implementing a method according to any one of claims 13 to 25.

27. A storage medium storing a computer program according to claim 26.