There are four 'different' voices: m1-4 and f1-4. The second line is a female voice talking loudly and quickly. The first line uses male voice 1, talking slowly and quietly. A couple of examples are given below:Įspeak -ven+m1 -s 100 -a 50 'hello world'Įspeak -ven+f1 -s 200 -a 100 'hello world' number of characters per second you can use the '-s' option, to adjust the volume use the '-a' option.

To change the voice/language or change between a male/female voice you can '-v' option, to adjust the speed i.e. The 'espeak' command is easy to use, to say 'hello' simply type: The final number is the volume, may of misread the range but it seems to be 0 to 200%? The pi is now ready to speak. To increase the volume use the following command: By default the sound level form the 3.5mm jack is quite low, i'm guessing its intended for headphones, therefore, a little on the quiet side for speakers. The final digit indicates the output: 0=auto, 1=3.5mm and 2=HDMI. Therefore, to configure the pi to use a specific audio output rather than the HDMI use the following command: By default the pi should auto-detect which output to use, but i found this didn't always work.

it doesn't support sound through the HDMI cable, need to use the 3.5mm audio out socket. Sudo apt-get install alsa-utils espeak python-espeakĪs my monitor uses a DVI input i.e. To install these software packages (i think : To allow the pi to play pre-recorded sounds, or general sound effects (.wav files), i also installed 'aplay'. There are a quite a few different command line tools and python libraries available, for no good reason i went for 'espeak'. Can you turn a Raspberry pi into Jarvis ( Link), well can you interact with a pi using voice commands? As with all projects started down a path and ended up with something else, discovered there was any easier solution, anyway.

Jarvis voice sounds wav