[asterisk-speech-rec] Min-Max Value Lumenvox Parameters

[Stephen Keller]

Hector Lopez wrote:
 
> Would you tell me the minimal and maximun values for parameters in
lumevox.cof
 
Actually, the parameters you're looking at are outdated, Hector. The
current version of the Engine (7.5.600) and above, when used with b16
and above of the connector bridge, support some new parameters.

I've copied over the contents of the new lumenvox.conf supplied with the
current connector bridge, which has some comments that are hopefully
more clear as to what each parameter does and how it can be set.

One thing worth noting is that snr_sensitivity and volume_sensitivity
are both set on scales from 0-100; I don't think most of the other
parameters have maximum values, as they are all time settings (but you
wouldn't want to set them very high).

Check out
http://www.lumenvox.com/helpProxy/index.aspx?project=speechengine&id=756
for information about the sensitivity settings.

We also have some recommended parameters at
http://www.lumenvox.com/helpProxy/index.aspx?project=speechengine&id=680

So basically I suggest you go to http://www.lumenvox.com/customers/ and
log in, grab the latest connector bridge and Engine, and install those
before you spend too much time tweaking the older parameters. The newer
VAD works much better is much easier to set.

; Signal-to-noise sensitivty
; Determines how much louder the speaker must be than the background
noise in
; order to trigger barge-in. The smaller this value, the easier it will 
; be to trigger barge-in.
vad_snr_sensitivity=50

; Volume sensitivity
; The volume required to trigger barge-in. The smaller the value, the
more 
; sensitive barge-in will be. This is primarily used to deal with poor
echo 
; cancellation. By setting this value higher (less sensitive) prompts
that are
; not properly cancelled will be less likely to falsely cancel barge-in.
vad_volume_sensitivity=50

; End-of-speech delay 
; This is the amount of time, specified in milliseconds, that the 
; Engine must detect silence after speech before it begins processing 
; the utterance. Set this value lower to capture short utterances such
as
; single words. Set it lower for longer utterances that are likely to
have many
; pauses in between speech, such as long digit strings.
vad_eos_delay=1250

; Wind-back length
; The length of audio to be wound back at the beginning of voice
activity. This
; is used primarily to counter instances where barge-in does not
accurately 
; capture the very start of speech. The resolution of this parameter 
; is 1/8 of a second.
vad_wind_back=750

; End of speech timeout
; This is the total amount of time to listen for speech after barge-in
has 
; been detected. This differs from the end-of-speech timeout as this is
; the total time a speaker has to speak, not the length of time between 
; individual words. This parameter should not usually need to be
changed.
end_of_speech_timeout=15000


Thanks,

Stephen Keller
LumenVox Support
P: 877-977-0707, just say "Support"
F: 858-707-7072
Support at LumenVox.com
www.LumenVox.com <http://www.lumenvox.com/> 

 <http://www.lumenvox.com/> 

Winner "Best Innovation in Speech Recognition"
AVIOS SpeechTEK Award