Arduino Text to Speech

Other Hardware 3rd Party Boards
1

Been wanting to try text to speech lately after I discovered the Windows Speech API for c++

Does arduino have anything similar? I found the emic 2 text to speech module and that looks awesome, does arduino have anything like that built in?

Also to use arduino and the emic 2 together, I have a speaker do I need an amplifier or is that optional, it sounds like the amplifier just boosts the sound?

2

This library might be of interest:

It's a bit tricky because that's the original repository but it's no longer actively developed. There are some forks that have had additional work done to them but I'm not sure which is the best. This one is probably worth a try:

3

pert:
This library might be of interest:
GitHub - going-digital/Talkie: Speech library for Arduino

It's a bit tricky because that's the original repository but it's no longer actively developed. There are some forks that have had additional work done to them but I'm not sure which is the best. This one is probably worth a try:
GitHub - PaulStoffregen/Talkie: Speech library for Arduino

Thanks, I actually got the emic 2 TTS module and have been playing around with it. It is amazing and exactly what I was looking for.

4

I was looking into this earlier in the year and for my implementation of "robo-clock" opted for pre-rendered segments of speech that I splice together on the fly.

Have routines to announce time, dates, days of the week, security messages, alarm times and even emotes depending day of week/temperature/humidity. The clock also keeps track of when it last detected movement and will greet anyone if it's been more than a certain amount of time since someone was around provided it's in a block of time that permits this feature (ie. not 3am)

All the sound fits on sd easily and has the advantage of being very clear and being a lot cheaper. Output was through an MCP4921 DAC.

I created a Windows app that reads a script and re-creates all the segments in batch so I can build with different voices and effects. The program creates the .wav files but more importantly creates a .h file with #define constants so segments can be used easily.

One segment of text may have 20 different ways of saying basically the same thing so it picks a random one.

If I can get fast reading from eprom I will have another bash at a dictionary approach.

5

Hoek:
I was looking into this earlier in the year and for my implementation of "robo-clock" opted for pre-rendered segments of speech that I splice together on the fly.

Have routines to announce time, dates, days of the week, security messages, alarm times and even emotes depending day of week/temperature/humidity. The clock also keeps track of when it last detected movement and will greet anyone if it's been more than a certain amount of time since someone was around provided it's in a block of time that permits this feature (ie. not 3am)

All the sound fits on sd easily and has the advantage of being very clear and being a lot cheaper. Output was through an MCP4921 DAC.

I created a Windows app that reads a script and re-creates all the segments in batch so I can build with different voices and effects. The program creates the .wav files but more importantly creates a .h file with #define constants so segments can be used easily.

One segment of text may have 20 different ways of saying basically the same thing so it picks a random one.

If I can get fast reading from eprom I will have another bash at a dictionary approach.

That's really cool! I can't wait until I can do stuff like that! I'm still messing around with arduino basics such as DC motors, ultrasonic sensors, servos etc.
Might I ask where you learned how to do all of that?