Problems with Mic picking up speaker
Printed From: MSSpeech-Forum
Category: Windows™ Speech Recognition Forums
Forum Name: New Users & General Questions
Forum Description: Ask questions, give and get answers.
URL: https://www.msspeech-forum.com/forum_posts.asp?TID=345
Printed Date: 27/Dec/2024 at 12:30am
Software Version: Web Wiz Forums 12.02 - http://www.webwizforums.com
Topic: Problems with Mic picking up speaker
Posted By: tod24
Subject: Problems with Mic picking up speaker
Date Posted: 08/Jan/2015 at 3:45am
|
Hi guys thanks for add to forums. Firstly I'm bit of a novice at all this voice recognition. My project is just to use MS VR in windows 7 by using Macros only' to execute sound files but I am running into problems with the mic picking up the sound been played at the same time, this then stop's the sound and sets of another command. Can I tweak windows 7 VR to not listen while playing. I have asked on another forums but since as this forum is for MSS maybe this is the better one to ask the question. Many thanks |
Replies:
Posted By: mmarkoe_admin
Date Posted: 08/Jan/2015 at 5:43am
Your problem is the microphone. Sound from your audio files are entering the audio stream and setting off your command macros. You need a good close talk microphone which will only pick up your voice and has background noise canceling. Check out the following web sites: 1. Speech Recognition Solutions (USA) 2. ITSpeaking (UK) I suggest you fly into a microphone that fits your price range and then send a query to the owner of the web site and ask if it should be noise canceling enough for what you wish to do. Marty |
tod24 wrote:Posted By: tod24
Date Posted: 09/Jan/2015 at 8:31am
|
Hmm not sure that will work marty The thing is I'm making a robot/Android kid using mics in the ears and a speaker inside the mouth so the speaker and mic will always be close together. So I think its going to be a software problem that will be my greatest hurdle. I will definatly upgrade the microphones though but they will have to be 2 of the tie-clip ones so I can fit inside the head and come out of the ears. So thats why i need to be able to speak commands from say a meter a way but then the voice that speaks back as you say enters the audio stream. I just need a process to kill this listen when in play mode ? Strange one I know lol but any ideas ? Tod |
Posted By: tod24
Date Posted: 09/Jan/2015 at 10:29am
| Hey Marty what's the last box used for in new speech macro's? What are optional command line arguments? |
Posted By: mmarkoe_admin
Date Posted: 09/Jan/2015 at 2:20pm
The first response to this posting made quickly from my phone said: "things like pauses." This is incorrect. Now at a computer and looking at http://www.mymsspeech.com/p/51/wsrmacros-the-users-guide - WSRMacros: The User's Guide the correct answer is: params -> ‘command line parameters’ to pass to the command (optional) <run command="notepad.exe" directory="c:\temp" params="c:\Users\Docs\MyFile.txt"/> Marty |
Posted By: mmarkoe_admin
Date Posted: 09/Jan/2015 at 2:25pm
Guess you could use a Pause command but bet this is going to be difficult if not impossible. Marty |
Posted By: tod24
Date Posted: 09/Jan/2015 at 3:58pm
|
Hmm just been thinking. What if i could use 3 commands in the macro
Not correct format but just an example: Mic off or Mic vol set 0 c:\sounds\hello.mp3 Mic on on mic set 5 Or how about I forget about trying to do this with MSR and use a 3rd party software to disable or change the mic volume to zero when a sound is been playing? |
Posted By: mmarkoe_admin
Date Posted: 10/Jan/2015 at 8:11am
I guess you could try this. The keyboard command to control the microphone On/Off is: CTRl + Windows Key You would need to set a pause equal to the MP3 audio file length before issuing the Mic On/Off command. Not sure how this would work out if there are possibilities of multiple audio files and playback lengths of different times.
What is MSR? Do you mean WSR, Windows Speech Recognition? Don't answer, it is just my way of letting you know the generally accepted terminology. Marty |
Posted By: tod24
Date Posted: 12/Jan/2015 at 5:38am
|
CTRl + Windows Key is not quite what I am looking for Today I have been playing around with Nircmd that switches off sound devices from the command prompt. I tested it out on the cd drive door opening and closing and worked fine then I tried the command Nircmd mutesubunitvolume Realtek High Definition Audio Microphone 1 It accepted the command but the microphone stayed on I tried the other Values 0 1 2 and still nothing WSR was still listening and the mic was still on. Any Idea where I am going wrong Marty? Ive also tried a small program called MicMute but all this does was create a hotkey same as WSR If they was a dos command for switching off Mic I could use this but cannot find anything like this
|
Posted By: mmarkoe_admin
Date Posted: 12/Jan/2015 at 8:53am
|
>> CTRl + Windows Key is not quite what I am looking for
The key combination works like an On/Off toggle switch. Keystrokes should be easily programmable. >> Any Idea where I am going wrong Marty? Yes, program the toggle keystrokes. Marty |
Posted By: tod24
Date Posted: 12/Jan/2015 at 11:44am
|
Erm.. not sure if you are being funny with me there mate! or I am just reading it out of context??? Thing is' if I was wanting to be sarcastic and a twat that's exactly how id' type it so??
Sorry I'm not your usual University / Computer programmer type but im not a dickhead!! I did say I was a novice didn't I ? Ok thanks for your advice before where you was quite helpful. But f**k knows; what the toggle keystrokes are or how to even program them?? Thanks Bye! Il ask somewhere else |
Posted By: mmarkoe_admin
Date Posted: 12/Jan/2015 at 12:25pm
|
Frankly, I am surprised why you thought I was being sarcastic. You had mentioned in the first posting that you're new to speech recognition. When you talked about creating a robot that would give voice responses to your commands, this implies a sophisticated level of programming knowledge.
Marty |
Posted By: tod24
Date Posted: 12/Jan/2015 at 1:02pm
|
Well maybe I took your reply wrong way ? but it looks like I have used up all my questions here anyhow' Sophisticated level of programming knowledge? Erm..not really' not the way I'm doing it and not if your using Macros and speech recognition. I think its piece of P**S The only problems are these small hurdles you find along the way eg: learning out how to disable the microphone with a command instead of manually pressing hot-keys and a big problem I had with multiple macros been heard, processed and running many of the sounds at the same time. I worked this one out myself. So I am half way there really. I will find out the command how to turn off and on the microphone from the command prompt. And I will then find out how to pause it or maybe think of another way to do this as my head is full of ideas. Then apart from that that's the main speech done. The next phase is the movement but to me that's the easy bit I'm a electrical engineer by trade so I know the basics. Servo motors in the neck assembly for pan and tilt also same with the eye movement all controlled by computer software. It will take a long time to get my project working perfect but I am willing to put the time i and learn it all. I'm moving on to the robotics forum now Thanks for your advice before Tod
|
Posted By: tod24
Date Posted: 12/Jan/2015 at 1:12pm
|
Marty, I think the bit where you repeat what I say and then answer in a short to the point answer without any other explanation of what you have just described comes across as arrogant' to the person that has asked the question and not as knowledgeable as you. Only saying' I bet I'm not the first person to take your answers the wrong way? lol cheers anyway Tod
|
Posted By: mmarkoe_admin
Date Posted: 12/Jan/2015 at 1:33pm
This is good forum protocol. It references the point a person mentions so there is no confusion.
Actually you are the first person to ever mention this. I'm truly sorry you are not getting the answers you would like to hear. My company has a programmer that has 8 years experience with WSR. I highly doubt he would take on the project you propose as experience (his and mine) understands the hurdles of sorting out speech from different people, coming from different directions and distances as well as background and reflected noise. Then you have the issue of the program selecting the correct response (audio file) to play. If someone asked me to invest time or money in this project, I would politely refuse. Best of luck. Your greatest revenge on my bluntness will be proving me wrong. 
Marty |