|
 |
Information |
The BMW Voice Recognition system
allows the driver to operate a number of the vehicle's functions simply by speaking.
When used in conjunction with the Multi Function Steering Wheel
(available on many recent models), you can also operate all principal functions and accessories. You can access phone functions, including recalling stored numbers and dialing, operate Navigation System functions, or take notes through
the built-in memo function.
Currently,
the size of the non-speaker-dependent vocabulary includes around 30 words, including numbers and commands. Spoken sequences of commands of up to five words and columns of numbers can be recognized with a high degree of accuracy.
You can create a telephone book with up to 40 numbers. Dialing is then simply a matter of speaking a name. Other normal telephone functions, such as repeat dialing and call hang-up are also voice activated.
|
While
not BMW spec yet, here is a sample
conversational voice recognition dialogue for interacting with an email reading
system. What we can expect very soon:
|
 | Driver: |
Mail |
|
 | Vehicle: |
Switching to mail. |
|
 | Driver: |
How many messages? |
|
 | Vehicle: |
There are two new messages. |
|
 | Driver: |
Read. |
|
 | Vehicle: |
[using text-to-speech]
"From: Andrea Palmorando
"Subject: RE: Evaluation
"Thanks for getting back to me with Jim's email address.
"As I told you, the CD went out via FedEx to you
yesterday afternoon…" |
| (Driver begins to decelerate rapidly...) |
|
 | Vehicle: |
(Emits "going quiet" noise) |
| (Car stops--perhaps at a stop light or intersection?) |
|
 | Vehicle: |
Shall I continue? |
|
 | Driver: |
Yes. |
|
 | Vehicle: |
"As I told you, the CD went out via FedEx to you
yesterday afternoon. I am going to send you a
license file via email. The license file will grant you access to the
software for which you are licensed.
"Please let me know if you have any questions and I
look forward to working with you." |
| (Car begins accelerating) |
|
 | Vehicle: |
(Emits "going quiet" noise) |
|
 | Driver: |
Read. |
|
 | Vehicle: |
(Driver overrides timeout period. Car returns to start of paragraph)
"Please let me know…" |
| (Driver
interrupts.) |
|
 | Driver: |
Next. |
|
 | Vehicle: |
"From: Bob Jims
"Subject: Invoice
"Attached is the MTD timesheet for our work so far in April.
"We are making good progress on phase two of the project..." |
| (Driver
interrupts.) |
|
 | Driver: |
Stop. |
| (Car stops speaking.) |
|
 | Driver: |
Rewind. |
| (Voice recognition mishears as "Delete".) |
|
 | Vehicle: |
Deleted message. |
|
 | Driver: |
What? |
|
 | Vehicle: |
The email message has been marked for deletion. |
| (Car begins decelerating.) |
|
 | Vehicle: |
(Emits "going quiet" noise) |
| (Car reaches cruising speed again.) |
|
 | Vehicle: |
(Emits "listening noise.") Email message number 2 has been marked for deletion. |
|
 | Driver: |
Undo. |
|
 | Vehicle: |
Email message restored. |
Dialogue
courtesy of i/net.
The dialogue demonstrates
that a conversational voice recognition interface provides additional safety and convenience features.
Here the system is paying attention to signals from the car (speed and acceleration) to know when to speak and when to go quiet. |
 |
Here's how it works: |
Voice recognition uses a
neural net to "learn" to recognize your voice. As you speak, the voice recognition software remembers the way you say each word. This customization allows voice recognition, even though everyone speaks with varying accents and inflection.
The voice commands
you use in your car are chosen from a fixed vocabulary and are passed on to the car telephone or navigation system via the telephone interface. The system gives acoustic feedback on everything recognized.

The system requires no lengthy voice recognition protocol and responds to a simple series of set voice commands that are not sensitive to the accent or dialect of the speaker. The voice control is a finite speech dialog system, which follows a predefined structure.
Faulty operation or error recognition can easily be corrected by simply repeating the desired command. The voice recognizer is resistant to stationary environmental noise.
|