Hi everyone! For my Arduino project, I need some Speech Recognition I can't find anything that fits all of my needs, please help me out here!
My criteria for such a system is this:
Must be SPEECH not VOICE recognition. Speech Recognition. recognizes anyone's voice, while Voice recognition is intended for only one persons voice. So no EasyVR shield, I guess (Unless there is a way to turn Voice Recognition into Sppech recognition. Could I set up my VR with such a generic voice that it recognizes anyones voice?)
Not limited. That Serial Speech Recognition module looks great, however all the commands must be separated into 5 groups and only 5 commands per group. A, I do not want to separate into 5 groups and B, I do not want to be limited to a total of 25 commands.
Bitvoicer does not work, because I have a MAC not a Windows. From my research, I cannot find a MAC version of Bitvoicer. If there is something out there for Mac, I am all ears!
The uSpeech library looked cool, but I was told that it's not that great in another forum. Is that true?
Any help would be AWESOME
That's amazing what you're asking here I had the same problem in the past on one of my project
you know how you can solve that? Use a camera, video, build a face recognition.
To build what you are asking "Speech recognition" you don't need just hardware but a software as well,
you need to record real time the sound of a from a person and process it thru a software that will compare the sound/freq/am/.etc with a database of sounds/voices from other people. But remember more than a person can have save voice or at least close to it. A person in his life can have different voices while naturally speaking, while cold, while smoking..etc
You project sounds fun and you can do it you need a software to build not a module, you just need a microfone, but I suggest you to use "voice recognition" with combination of "face/person recognition"
Please stick to asking questions you do not know nearly enough even to begin to answer them.
What kind of questions? btw this is not my topic i suggested face recognition, I worked on face recognition at University it's easier to do that than speech (personal voice) recognition.
The sort of very low level questions that you usually ask and fail to understand any of the answers you get.
btw this is not my topic
I know, you just jumped in with totally useless advice. The OP was asking about voice recognition and true to form you failed to see how your comment was irrelevant.
I worked on face recognition at University it's easier to do that than speech (personal voice) recognition.
I find it hard to believe that you went to University.
You obviously did not understand what you were doing.
Ok, so let's get back on task please. Let's all answer the ORIGINAL QUESTION
mart256:
You need a pc, not an arduino.
So it seems like a computer is the best way to go when doing Speech Recognition. So something like BitVoicer. As I said originally, I have a MAC and there is no mac version of BitVoicer that I can find. So, following that logic,
What can I do with my mac to do Speech Recognition and signal the Arduino? Is there a Mac version of BitVoicer? Is there anything LIKE BitVoicer that could work?
WesleyRTech:
Ok, so let's get back on task please. Let's all answer the ORIGINAL QUESTION
So it seems like a computer is the best way to go when doing Speech Recognition. So something like BitVoicer. As I said originally, I have a MAC and there is no mac version of BitVoicer that I can find. So, following that logic,
What can I do with my mac to do Speech Recognition and signal the Arduino? Is there a Mac version of BitVoicer? Is there anything LIKE BitVoicer that could work?
Virtual machine? Dual boot? There are many ways to overcome the os limits.
No, you look at the manual again. Though it does require a PC and extra license fee.
Speaker Independent Custom V
ocabularies
The set of built
in Speaker Independent recognition vocabularies can be expanded with custom grammars,
that you can create with the QuickT2SI
TM
tool (a separate license is required to use the software).
When you create a QuickT2SI
TM
project, you are presented with a list of words or short phrases (also called
“commands”) and an optional trigger word/phrase. The so
called “trigger” is a special set that contains only
one word or phrase, with an improved recognition performance, that i
s used as an entry point for any vocal
interaction with a device that is continuously listening to the user’s voice.
If you need to use a trigger word, it is important to carefully choose it so that it has good performance, with
very few unintended activa
tions and a high recognition rate. When the user says the trigger word followed by
a command, the system can discard unintended activations when the trigger is not followed by a command
within a short amount of time (usually around 3 seconds). Moreover, th
ere is only one trigger word to listen
to, instead of a list of several commands, so the chance to pick up a random command from background
noise or talk is also lower, when using a trigger word.
For assistance on using the QuickT2SI
™
Software, please ref
er to the software help file
EasyVR comes with free software for speaker-dependant commands.
EasyVR Commander.
You can add/record/train upto about 30 commands.
I am using that shield on an RF remote control for my home lighting.
e.g. "Lights_on" only reacts to my voice, not to the voice of other members of the family.
EasyVR has a steep learning curve. Took me days to get it to do what I wanted.
Leo..
Domino60:
wow that's really offencive, didn't expected such thing from a Brattain Member
shame of you.
I agree. That is offensive. Why don't we all stop hating each other, and start showing some respect? Or are you all too self-rite ours to focus on the question originally posed?
