Direct voice input (DVI) allows a single crew member to interface with an aircraft through speech commands in high workload situations. Speech recognition systems have achieved over 95% accuracy rates but decrease to less than 80% under high g-forces. Voice control is not suitable for time-critical systems. DVI has been applied for non-critical tasks like radio tuning and map display but not weapons firing due to response time and recognition problems, especially with similar sounding words. Multi-function displays like Collins Aerospace's MFD-268 provide enhanced situational awareness through high resolution, sunlight readable screens that can display maps, video and more in both landscape and portrait orientations.

1. Unit 3
Topdays’ Topics
DVI
MFD

2. Direct Voice
Input
DVI

3. • It’s a kind of Interface between the crew and a/c in
high work load situation in single crew member a/c
• F-16 routinely achieved 95% correct word recognition
and reduced to less than 80% under high g
conditions
• Voice control is not suitable for time critical system.

4. • Recognition
• Speech recognition
• Synthesis
• Mature and proven performer

5. • Applied for non critical task such as requesting system
• Status,
• tuning radios,
• And requesting maps to be displayed on a CRT
• Not been used for urgent Inputs or critical task such
as firing weapons.

6. • The words in the vocabulary are limited.
• Generating templates are time consuming.
• Microphone have the same electrical characteristics
as the flight microphone
• Difficult to stimulate the stress artificially
• Speaker independent Speech Recognition
requires large amount of memory ,and low signal
processing

7. • Correct recognition
• At a rate in the mid 90 % and decreases to less than 80%
for a 9-g load on the pilot
• Confusion with a rhyming word
• And rejection
• The response time and the recognition problems can
be reduced by
• Syntactical rules and logic

9. • Node 1 is the master node
• Contains all the word that are permitted to
be the first word in a spoken sequence
 Node 2 to 4 are secondary nodes contains second
words on the set of words based on the initial word
and the syntactical rules.
 Ex: Comm1 refers radio ,so logic dictate that the words
which follow are numbers representing the frequency to
which the radio should be tuned

10. Working of speech recognition
system
*It uses hidden Markov modelling (HMM)
*2 pattern modes:
-Acoustical modelling
-Language modelling

11. Acoustical modelling
• Phonemes: a phoneme is a unit of sound that distinguishes one word
from another in a particular language
• In this modelling, uttered words are processed into phonemes.
• English language is made with 45-50 phonemes.
• Phonemes are converted to digital format
• This digital format or pattern is matched with a

13. Commands

14. MULTI FUNCTION
DISPLAYS (MFD)

15. MFD
• Collins Aerospace’s MFD-268
• Used in rugged military helicopters
and fixed wing a/cs
• Resolution 1024*768
• AMLCD display
• Offers landscape and portrait
configurations
• Provides enhanced situational
awareness all phases of flight

16. Features
and benefits
High quality
Sunlight readable and avionics grade
High brightness at broad dimming range
NVIS compatible
Available as video monitor
Multiple video formats FLIR/ map