Thanks Mike, george, linnix,
I have checked out the links & will investigate them furthur. Actually my background is in programming so I kind of feel like a fish out of water on this subject. (Although it may sound like a school project, its actually for end consumers)
I am familiar with PWM techniques in general. (it sounds like this is the way to go)
Once I land on a chipset & receive a development kit I think I will feel more comfortable from there.
Let me elaborate a little more on the project:
- The toy is meant to play musical songs with a digital file running simaltaneously in the background.
- Fairly high quality sounds are required (for a toy) minimum 8khz 8bit mono.
- Melodies must play for 2 to 5 minutes in the foreground while digital sounds can repeat in the background(at least 20 seconds)
- oboard EEPROM for code as well as data is preferred
- However, this is meant to be an upscale item, so a chip/chipset costing would not be out of the question. (Quality is mandatory)
I am figuring on at least 20 secs of digital sounds (8khz 8bit mono) without compression off the top of my head probably comes to ~150k. Plus the 2 minutes of melodies which shouldn't take nearly that much.
At this much memory, any thoughts on hardware decompression, software decompression or going without decompression?
I do have some background in this field, as I wrote a custom Sound Blaster Driver years ago. I would prefer a chip capable of 2 seperate channels(mono) of audio as opposed to mixing the channels into one.
I really appreciate the help in making this selection, I feel a little overwhelmed with the choices...
-Tim