speech recognition and synthesis
#1
Posted 17 January 2012 - 10:44 AM
#2
Posted 17 January 2012 - 05:34 PM
I think the Kinect is the best way to go for this. There is a sample out there already - I'd try Googling "Simon Kinect Netduino". You really need a PC to do this as the processing power and libraries required are both hefty.heyy !!
i am planning to start my semester project using netduino (tired of programming PIC )
my project includes speech recognition and synthesis ,, and i seriously dont want to buy full KINECT just for that ,,
is there any simple way for speech recognition ( using kinect's library and a simple microphone at adc port)
its a 32 bit 48 MHZ ,, so would be lame if i had to include other IC for speech recognition processing
There are some speech synthesis chips on Sparkfuns site but the output is very 1980 retro game style.
Hope this helps and welcome to the community.
Regards,
Mark
To be or not to be = 0xFF
Blogging about Netduino, .NET, STM8S and STM32 and generally waffling on about life
Follow @nevynuk on Twitter
#3
Posted 18 January 2012 - 06:42 AM
#4
Posted 20 January 2012 - 08:40 AM
tnks mark for the reply
i was thinking of capturing voice and send it to a cloud app for recognition(for project will dewploy on my pc
but tue question is is there any library to send voice stream to cloud app (i can send voice packets but can anyone auggest for stream
Hello ash_36_ohms,
I am in the process of planning a something similar. My notion is go to with the Dragon Naturally Speaking SDK, as I've some experience with their products. If the API is anything as good as their desktop software, I'll be impressed with that as well as their stand-alone products. It will come at a price, though - at present I've not yet learned how much it is.
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users