node-red-contrib-wavenet

A Node-RED custom node to convert text to speech using Google Wavenet voice.

Update: Now with 60+ voices with options to adjust speaking rate, pitch, volume, sample rate and audio encoding.

Latest languages added:

• Added Arabic voices
• Added Danish voices

Wavenet voice represents a new way of creating synthetic speech, using a WaveNet model, the same technology used to produce speech for Google Assistant, Google Search, and Google Translate.

Takes a string as an input. Returns a base64 encoded array buffer. Attach a file node to save it as an audio file on your disk.

Sign in to Google Developer Console to get a Speech API key.

Inputs

Payload (string)

• Use SSML elements in your text to give it more personality. You can use all elements found here.

Voice (string)

• Choose a specific voice or random

Speaking rate (number | 0.25 to 4.0)

• In the range of 0.25 to 4.0. 1.0 is the normal native speed. 2.0 is twice as fast, and 0.5 is half as fast.

Pitch (number | -20 to +20)

• In the range -20.0 to 20.0. 20 means increase 20 semitones from the original pitch. -20 means decrease 20 semitones from the original pitch.

Volume (number | -6 to +6)

• A value of -6.0 (dB) will play at approximately half the amplitude and a value of +6.0 (dB) will play at approximately twice the amplitude of the normal native signal.

Sample rate (number)

• The synthesis sample rate (in hertz). When this is specified, the synthesizer will honor this request by converting to the desired sample rate. Default is 24000.

Audio encoding (string)

• The encoding format of the audio byte stream can be MP3, LINEAR16 or OGG_OPUS

Output

payload (string | base64)

• A base64-encoded string that is wrapped in containers according to the encoding specified.