VC-to-DTMF interfacing system and method

• US 7,092,515 B2
• Filed: 09/06/2003
• Issued: 08/15/2006
• Est. Priority Date: 09/06/2003
• Status: Active Grant

• Current Assignee: Applied Voice & Speech Technologies Incorporated (Open Text Corporation)
• Original Assignee: Applied Voice & Speech Technologies Incorporated (Open Text Corporation)
• Inventors: Gierachf, Karl
• Primary Examiner(s): Tran, Sinh
• Assistant Examiner(s): Briney, III, Walter F
• Application Number:
  US10/655,881
• Publication Number:
  US 20050053230A1
• Time in Patent Office: 1,074 Days
• Field of Search: 379/406.01, 379 8801- 8804, 379 8817- 8818, 379 8822- 8823
• US Class Current: 379/406.01
• CPC Class Codes: H04M 9/082 using echo cancellers echo cancellers per se H04B3/23

Abstract

A voice command (VC)-to-DTMF interfacing system that allows existing DTMF-driven systems to be interacted with by a caller using voice commands. The VC-to-DTMF interfacing system translates the voice command from a caller into the DTMF codes which are then played through a port for receipt by the DTMF-driven system. In a first mode, the voice commands from the caller and the pre-recorded messages of the DTMF-driven system are echo cancelled. In a second mode, a voice message is directly streamed to the DTMF-driven system through an enabled port patch.

Claims

1. A voice command (VC) to dual tone multi-frequency (DTMF) interfacing system comprising:

a first echo canceller for echo canceling a received voice command from a caller in a first mode;

a translator for translating the echo cancelled voice command into a DTMF code for use by a DTMF-driven system;

a second echo canceller for echo canceling audio output sent to the caller from the DTMF-driven system;

a first port for receiving a call from the caller;

a second port for sending the DTMF code to and receiving the audio output from the DTMF-driven system;

a port patch for connecting audio from the first port directly to the second port in a second mode; and

a tone detector for detecting a tone, in the first mode, from the DTMF-driven system to switch to the second mode.

2. The system according to claim 1, further comprising:

a DTMF digit detector for detecting a predetermined DTMF digit, in the second mode, from the caller to switch to the first mode.

3. The system according to claim 1, further comprising:

a DTMF digit detector for detecting a predetermined DTMF digit, in the second mode, from the caller to switch to the first mode and to forward DTMF codes to the DTMF-driven system when voice commands are not used.

4. The system according to claim 1, wherein:

during the second mode, the audio is a voice message to be stored in a voice mailbox of the DTMF-driven system; and

during the first mode, the voice message stored in the voice mailbox can be retrieved.

5. The system according to claim 1, further comprising:

an automatic speech recognition module for recognizing the voice command;

wherein the translator includes;

a plurality of audio files, each audio file corresponding to a DTMF tone wherein a distinct ordered combination of the plurality of audio files is associated with each voice command.

6. The system according to claim 5, wherein the translator further comprises:

a DTMF audio file player.

7. A method of interacting with a dual tone multi-frequency (DTMF) driven system with voice commands comprising steps of:

echo canceling a received voice command from a caller in a first mode;

translating the echo cancelled voice command into a DTMF code for use by the DTMF-driven system;

echo canceling audio output sent to the caller from the DTMF-driven system;

receiving a call from the caller at a first port;

sending the DTMF code to and receiving the audio output from the DTMF-driven system at a second port;

enabling a port patch for connecting audio from the first port directly to the second port in a second mode;

detecting a tone from the DTMF-driven system, in the first mode, to switch to the second mode; and

enabling the port patch, in response to the detecting step.

8. The method according to claim 7, further comprising the steps of:

detecting a predetermined DTMF digit, in the second mode, from the caller to disable the port patch; and

disabling the port patch, in response to detection of the predetermined DTMF digit.

9. The method according to claim 7, further comprising the steps of:

detecting a predetermined keyword, in the second mode, from the caller to disable the port patch; and

disabling the port patch, in response to detection of the predetermined DTMF digit.

10. The method according to claim 9, wherein:

during the second mode, the audio is a voice message to be stored in a voice mailbox of the DTMF-driven system; and

during the first mode, the voice message stored in the voice mailbox can be retrieved.

11. The method according to claim 7, further comprising the step of:

automatically recognizing the voice command;

wherein the translating step includes;

determining an ordered set of DTMF codes associated with the voice command, where each DTMF code has a one-to-one correspondence with an audio file containing an audio representation of that DTMF code.

12. The method according to claim 11, wherein the translating step further comprises the step of:

playing the ordered set of DTMF audio files through a port connected to the DTMF-driven system.

13. A method of interacting with a dual tone multi-frequency (DTMF) driven system with voice commands, the method comprising steps of:

receiving at a first port audio input from a user, the audio input comprising the voice commands;

translating the voice commands into corresponding DTMF codes for use by the DTMF driven system;

in response to the step of receiving, connecting to the DTMF driven system through a second port to receive audio output of the DTMF driven system and send the corresponding DTMF codes to the DTMF driven system;

sending the audio output of the DTMF driven system to the user through the first port;

detecting a tone from the DTMF driven system, the tone signaling the user to record a message;

in response to detection of the tone, enabling a port patch for connecting audio input of the user from the first port directly to the second port.