Method and apparatus for extracting voiced telephone numbers and email addresses from voice mail messages
AssigneeMitel Knowledge Corporation
InventorsSteve Horvath, Tonis Kasvand
Life of this patent
Estimating the term…
Loading patent…
AssigneeMitel Knowledge Corporation
InventorsSteve Horvath, Tonis Kasvand
Estimating the term…
The present invention relates to a system for automatically identifying enunciation of telephone numbers and email addresses in voice mail messages for presentation to a user via either voice or text, without the user being required to replay the message. Furthermore, according to the invention the extracted telephone number or email addresses may be forwarded to an application for further processing (e.g. to dial the extracted number, add the email address and/or telephone number to a contacts database, etc.) The user may specify text strings to be searched (other than telephone numbers and email addresses). Thus, a user can specify certain words such as “extension”, “email”, etc., in order to qualify the extracted number or email address information, or words such as “urgently” or “immediately” to identify important voice mail messages.
A system for extracting information from a voicemail message, comprising:
A method of extracting information from a voicemail message, comprising:
Dependencies are inferred from the claim text.
Drawings are temporarily unavailable. Open the PDF at USPTO
From the US filing history. Foreign family members aren’t listed.
Voice mail systems: message receiving aspects
Message receiving aspects: message type or catagory, e.g. priority, indication
Electronic components, circuits, software, systems or apparatus used in telephone systems: using speech recognition
Related to audio recordings in general: management of recordings
Term ended 23 Nov 2022 (estimated)
The present invention relates in general to voice mail systems and more specifically to a method and apparatus for identifying and extracting voiced telephone numbers and email addresses from recorded voice mail messages.
A common complaint among users of voice mail systems relates to difficulties in extracting and transcribing important information from voice mail messages for later use. As an example, it is common for a caller whose call is answered by voice mail to leave a voice mail message that includes a telephone number or email address at which the caller may be contacted. Unless the caller enunciates the telephone number or email address very slowly, the voice mail recipient may be forced to replay all or a portion of the voice mail message in order to transcribe the telephone number or email address for later use. Most callers find very slow enunciation to be awkward and unnatural, with the result that the majority of such voice mail messages must be replayed once or even several times in order for the voice mail recipient to transcribe the information. Furthermore, the process for replaying voice mail messages is often complex (e.g. requiring memorization of keys on a telephone keypad which have been assigned for the replay function). The process can also be tedious (e.g. some systems require that the entire message be replayed in order to extract only the portion relating to the telephone number or email address).
According to the present invention, a system is provided for automatically identifying enunciation of telephone numbers and email addresses in voice mail messages for presentation to a user via either a voice or text, without the user being required to replay the message. Furthermore, according to the invention the extracted telephone number or email addresses may be forwarded to an application for further processing (e.g. to dial the extracted number, add the email address and/or telephone number to a contacts database, etc.)
A description of a preferred embodiment of the present invention is provided herein below with reference to the accompanying drawings in which:
FIG. 1 is a schematic representation of a PBX voice mail system incorporating telephone number extraction in accordance with a preferred embodiment of the present invention;
FIG. 2 is a graphical user interface showing an example of the manner in which extracted information is presented to the user according to the invention;
With reference to FIG. 1, a server based PBX is shown (i.e. PBX) comprising a server 1, to which a plurality of communication endpoints are connected such as telephone 3 and PC 5. The telephone 3 can be a digital set, POTS set, a PC phone or LAN/IP phone. For ease of illustration and description, only two endpoints are depicted in FIG. 1. In practice, a multiplicity of endpoint devices is provided. The server 1 establishes communication between the endpoints under control of call processing software 7, in a well-known manner.
According to the invention, a voice number extraction voicemail application 9 uses an audio file 11, knowledge database 12, and sound database 13, the operations of which are set forth in greater detail below. The telephone 3 includes a plurality of feature keys for executing commands in connection with the voice number extraction application 9. Specifically, a “Listen to Message” key 15 is provided which, when pressed by a telephone user causes DTMF digits (or other signaling) to be sent by the telephone 3 to the call processing software 7. The call processing software 7 then activates the application 9 to retrieve voice mail messages from audio file 11.
In response to pressing a “Find Number” key 17 while listening to a voice mail message, DTMF digits (or other signaling) are sent by the telephone 3 to the call processing software 7. A keypad access code, or soft key on telephone 3, or a GUI button or menu item on PC 5 may replace the key 17. The call processing software 7 then causes the application 9 to launch a voice recognition application (e.g. Speak@Ease by Mitel Corporation) with instructions to search for a number in the retrieved voice mail message, as described in greater detail below. When the voice recognition application finds a number, it is presented to the user at telephone 3 or PC 5 via voice or textual display (provided the phone 3 supports textual display). It is contemplated that the extracted phone number may be validated against the phone number extracted from Caller ID data that may be stored in the voice message. Upon extraction of a number, the application 9 generates a user prompt for further instructions. The user prompt can be either voice or text. In response to the prompt, the user can select from one of several feature keys (or access codes, or soft keys or GUI buttons or menu items) identified in FIG. 1 as “Accept Number” key 19, “Continue Search” key 21, or “Abandon Search” key 23.