rhasspy-Azure

Simple shell script to use Azure Speech service (MS's cloud TTS service) in Rhasspy

Based on the jarvis_says.sh script from tschmidty69's Home Assistant config (his was for Snips and used Amazon Polly TTS) with help from this thread on the Rhasspy forums and also inspired by rhasspy-IBMWatson

Runs only on Rhasspy 2.5.11 or later

Preparation

1. Create an account on Azure and register for the Azure free account here: https://azure.microsoft.com/en-us/free/

• Create resource -> Speech
• Choose subscription and resource group
• Use the closest region
• You can change the service name
• Choose pricing tier to Free F0
• Keep the defaults otherwise

The free plan includes 0.5M characters/month which is plenty for personal home use.

2. In your dashboard find your service
3. Note your APIKEY and URL

Manual Installation

1. Copy the rhasspy_azure_says.sh file inside your rhasspy profile folder (rhasspy/profiles/en/)
2. Update the apikey, url, voice and cache variables

• apikey = Your Azure apikey
• tokenurl = Your endpoint url
• voice, lang, gender = The Azure speech voice to use (see https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/language-support?tabs=tts#tabpanel_1_tts for a complete list) (Example:voice en-GB-BellaNeural, lang en-GB, gender Female)
• OutputFormat = Choose non-streaming PCM audio formats (see https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/rest-text-to-speech?tabs=nonstreaming#tabpanel_1_nonstreaming)
• host = Region url (see https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/rest-text-to-speech?tabs=streaming#convert-text-to-speech)
• useragent = The application name. (Example:myApp/1.0.0)
• cache = the path to the cache folder:

Installation	Cache path
Docker	/profiles/en/tts/cache/
Non-docker	/home/${USER}/.config/rhasspy/profiles/en/tts/cache/

3. Make the script executable: chmod a+x /home/${USER}/.config/rhasspy/profiles/en/rhasspy_azure_says.sh (you might have to run with sudo)
4. Update the Rhasspy TTS settings to Local Command and set Say Program to:

Installation	Say program path
Docker	/profiles/en/rhasspy_azure_says.sh
Non-docker	/home/${USER}/.config/rhasspy/profiles/en/rhasspy_azure_says.sh