Function for Text to Speech conversion

This function converts text to speech using the alpine flite package.

It is installed in the runtime image using a custom Dockerfile which is only slightly different compared to the default one i.e. one line has been added to the runtime stage of the Docker build - RUN apk add --no-cache flite. This installs the flite package into the Docker image itself

The Go function

• uses os.Command to invoke the flite binary
• which converts the text into speech and saves it into /tmp/output.wav (this is deleted before the function exits)
• the file is then read and the raw bytes are simply returned to the caller

Pre-requisites

• Start by cloning this repository
• Configure your functions development environment
• Get latest Fn CLI - curl -LSs https://raw.githubusercontent.com/fnproject/cli/master/install | sh
• Switch to right context using fn use context <context name>

Create an app

fn create app text2speech --annotation oracle.com/oci/subnetIds=<SUBNETS>

Deploy to Oracle Functions

fn -v deploy --app text2speech

Invoke (and listen..)

echo -n 'Hope you enjoyed this text to speech example using Oracle Functions' | fn invoke text2speech talk > op.wav && afplay op.wav

echo -n 'That was dope!' | fn invoke text2speech talk > op.wav && afplay op.wav

^^ (these work seamlessly on Mac)