Text to speech

Text to speech in a cloud or local. Runs on windows, linux and OSX. It can play mp3/wav too.

Current Release
5.3.1
Developer
bluefox
License
MIT

This adapter uses Sentry libraries to automatically report exceptions and code errors to the developers. For more details and for information how to disable the error reporting see Sentry-Plugin Documentation! Sentry reporting is used starting with js-controller 3.0.

SayIt Adapter can convert text to speech and play it on some device.

Configuration

Actually, the following outputs are supported:

  • Browser - browser will play the text within opened iobroker.vis page. It is supported almost by every desktop browser and by few mobile Browsers.

  • Home24- MediaPlayer - the text will be sent and played to the Android device with Home24 - MediaPlayer installed. For this will be used build in Android text to speech engine. The port cannot be changed and set to 50000.

  • Home24 - MediaPlayer and FTP Server - the text will be sent and played on the Android device with Home24 - MediaPlayer installed. For this will be used the Google text to speech engine. Generated mp3 file will be copied over FTP to an Android device and played with Home24 - MediaPlayer. Both apps have to have same home directories. (E.g., root directory of "sd card").

  • System - the text will be played by OS, where the ioBroker adapter runs. Following OS are supported: Windows, linux, Mac OSx.

  • Windows engine - the text will be played by windows, where the sayIt adapter runs. For this will be used the Windows text to speech engine, that should be preconfigured by user. You can check here how to set up it.

  • Sonos - play text on the Sonos device. Be sure the Web Adapter is enabled. It is required to enable SONOS to read the generated mp3 files.

  • Heos - play text on HEOS device. Be sure the Web Adapter is enabled. It is required to enable HEOS to read the generated mp3 files.

  • Chromecast - play text on a Chromecast device.

  • MPD - play text on Music Player Daemon. Use only http for web adapter, don't use https.

To enable the text to speech on RaspberryPI or linux system call one time following command sudo apt-get -y install mpg321 to install mpg321.

The mp3/wav files can be played to by writing its name into the object. (e.g. /vis.0/main/img/door-bell.mp3)

The file must be first loaded.

TTS engines

Online:

  • Google: English, German, Russian, Italian, Spanish, French;
  • Yandex: Russian To use Yandex voices you must request the API key here: https://tech.yandex.ru/speechkit/cloud/doc/dg/concepts/About-docpage/. [This service will be disabled 1st of Jan 2019 and replaced by Yandex.cloud] To use Yandex.cloud you should register here: [https://cloud.yandex.ru/], install SpeechKIT API in the Cloud and get Auth Token and Folder ID as described in API instructions.
  • FreeTTS: more than 400 voices in more than 75 languages, provided by https://freetts.org. The language is not selected by the engine, but by the voice, e.g. de-DE-KatjaNeural. An API key from the pricing page is required. The free tier of the service appends a spoken "generated with freeTTS.org" to every text, so it cannot be used for announcements.
  • Cloud: To use Cloud voices you need configured and running cloud adapter or enter app-key directly in settings
  • Amazon Web Services Polly: To use AWS Polly voices, you need to create an access key and secret key here. The Amazon documentation can you find here.

Offline:

  • PicoTTS (linux only): English, German, Italian, Spanish, French; For PicoTTS it is necessary to install the following packages: libttspico-utils and lame. Installation command: sudo apt-get install libttspico-utils lame

  • Coqui TTS: English, German, Spanish, French, Dutch, Japanese, Chinese; For instructions how to use go to the official documentation

Test of the text generation

With the button Test text generation on the tab Engine you can check if the selected engine is configured properly.

The field Test text is prefilled with a test sentence in the language of the selected engine and can be changed as desired. If the field is empty, the same sentence will be used. The text will be generated with the settings of the dialog, so they must not be saved before the test.

The text will be only generated and not played, so the test works even if no player is available. The dialog shows the size and the duration of the generated file and tries to open it in a new browser tab, so you can listen to it. For that the browser must allow pop-ups for the ioBroker admin.

To test the generation and the playback together, use the button Test on the tab Player.

Cloud and Amazon Web Services Polly text formatting

You can format your text with Speech Synthesis Markup Language.

Most useful features:

  • <break time="3s"/>- make a pause for x seconds (max 10 seconds).
  • <emphasis> big </emphasis> - make an emphasis on some word.
  • <prosody volume="+6dB" rate="90%">I am speaking this</prosody> - control speed and volume parameters.
  • <say-as interpret-as="digits">12345</say-as> - say every digit separately.

More info.

System command

If you have some program, that can play audio files locally or somewhere else, you can write this command here. E.g.

myCustomPlayer --option

If System output is selected, the sayit adapter will execute the following command on a local system:

myCustomPlayer --option /opt/iobroker/node_modules/iobroker.sayit/say.mp3

If file name must stay somewhere in the middle, you can use %s to specify where the file name must be placed:

myCustomPlayer --option "%s" > /dev/null

sayIt will make myCustomPlayer --option "/opt/iobroker/node_modules/iobroker.sayit/say.mp3" > /dev/null from it.

Usage

SayIt adapter cannot be used alone. It must be controlled from javascript adapter or from "vis" with specific widget. After creation of adapter instance, you can find the following objects:

  • sayit.N.tts.text: Phrase to be spoken.
  • sayit.N.tts.volume: volume which will be used by playing of the phrase.
  • sayit.N.tts.playing: true if a text is now playing and false if not. Supported only for "windows" and "system" play mode.
  • sayit.N.tts.cachetext: Phrase to be cached, and then it can be used without an internet. E.g., you can enter here manually "No internet" and if ping to google.com is negative, write "No internet" to "tts.text" and it will be pronounced. Of course, cache must be enabled.

State tts.text supports extended syntax, so the language/engine and volume can be defined together with text. It is used to enable multi-language text2speech engines. E.g., if adapter has engine Google-english, it is possible with phrase de:Sag es to force using Google-Deutsch speech engine.

With ru;75;Погода хорошая we can force using russian language and volume 75%.

You can specify the volume of announcement in percent from current or given volume (not from maximal). E.g., if command is de;75;Gutes Wetterand "announce volume" is 50%, the announcement will be played with volume 38% from 100% possible.

The system command to play the mp3 file can be specified too. If you leave it blank, the default settings will be used: windows - cmdmp3.exe, OSX - /usr/bin/afplay, linux - mpg321 or omxplayer (recommended).

To install omxplayer write sudo apt-get install omxplayer or write sudo apt-get install mpg321 to install mpg321.

Note: The default announce selection will be possible only after the start of the instance.

With sendTo you can send more different parameters, like sonosDevice or browserInstance:

sendTo('sayit.0', 'say', {
    text: 'Hello',
    sonosDevice: 'Wohnzimmer', // optional, if not defined, the device from configuration will be used
    engine: 'Google-de', // optional, if not defined, the device from configuration will be used
    type: 'sonos', // optional, if not defined, the device from configuration will be used
    volume: 20, // optional, if not defined, the device from configuration will be used
});

Priorities

To immediately pronounce the text despite the queued texts, you have two possibilities:

  • place "!" as a first character in text, so this text will be pronounced immediately after the current one.
  • write true into "tts.clearQueue" state and the queue will be cleared. After that you can write a new text into tts.text, but all queued texts are thrown away.

Engines

The following values for engines are possible:

Google

  • en - English
  • de - Deutsch
  • pl - Polski
  • ru - Русский
  • uk - український
  • it - Italiano
  • es - Espaniol
  • fr - Français
  • nl - Nederlands
  • zh-CN - 简体中文
  • pt - Português

FreeTTS

  • freeTTS - the voice, the speed and the pitch are taken from the instance configuration

The language is a part of the voice, so only one engine name exists. The voice, e.g. de-DE-KatjaNeural, is selected in the configuration dialog and the list of all available voices is read from the service. A request is limited to 10000 characters, so longer texts are split and said one after another.

Yandex

  • ru_YA:Yandex - Русский
  • ru_YA_CLOUD:Yandex Cloud - Русский [Yandex.Cloud API generates files in OGG format. To play ogg files on linux mplayer should be installed and selected as system player]

Amazon polly via cloud

  • ru-RU_CLOUD_Female - Русский - Татьяна
  • ru-RU_CLOUD_Male - Русский - Максим
  • de-DE_CLOUD_Female - Deutsch - Marlene
  • de-DE_CLOUD_Male - Deutsch - Hans
  • de-DE_CLOUD_Female_Vicki - Deutsch - Vicki
  • de-DE_CLOUD_Male_Daniel - Deutsch - Daniel
  • de-AT_CLOUD_Female_Hannah - Österreich - Hannah
  • en-US_CLOUD_Female - en-US - Female - Salli
  • en-US_CLOUD_Male - en-US - Male - Joey
  • da-DK_CLOUD_Female - da-DK - Female - Naja
  • da-DK_CLOUD_Male - da-DK - Male - Mads
  • en-AU_CLOUD_Female - en-AU - Female - Nicole
  • en-AU_CLOUD_Male - en-AU - Male - Russell
  • en-GB_CLOUD_Female_Amy - en-GB - Female - Amy
  • en-GB_CLOUD_Male - en-GB - Male - Brian
  • en-GB_CLOUD_Female_Emma - en-GB - Female - Emma
  • en-GB-WLS_CLOUD_Female - en-GB-WLS - Female - Gwyneth
  • en-GB-WLS_CLOUD_Male - en-GB-WLS - Male - Geraint
  • cy-GB_CLOUD_Female - cy-GB - Female - Gwyneth
  • cy-GB_CLOUD_Male - cy-GB - Male - Geraint
  • en-IN_CLOUD_Female - en-IN - Female - Raveena
  • en-US_CLOUD_Male_Chipmunk - en-US - Male - Chipmunk
  • en-US_CLOUD_Male_Eric - en-US - Male - Eric
  • en-US_CLOUD_Female_Ivy - en-US - Female - Ivy
  • en-US_CLOUD_Female_Jennifer - en-US - Female - Jennifer
  • en-US_CLOUD_Male_Justin - en-US - Male - Justin
  • en-US_CLOUD_Female_Kendra - en-US - Female - Kendra
  • en-US_CLOUD_Female_Kimberly - en-US - Female - Kimberly
  • es-ES_CLOUD_Female - es-ES - Female - Conchita
  • es-ES_CLOUD_Male - es-ES - Male - Enrique
  • es-US_CLOUD_Female - es-US - Female - Penelope
  • es-US_CLOUD_Male - es-US - Male - Miguel
  • fr-CA_CLOUD_Female - fr-CA - Female - Chantal
  • fr-FR_CLOUD_Female - fr-FR - Female - Celine
  • fr-FR_CLOUD_Male - fr-FR - Male - Mathieu
  • is-IS_CLOUD_Female - is-IS - Female - Dora
  • is-IS_CLOUD_Male - is-IS - Male - Karl
  • it-IT_CLOUD_Female - it-IT - Female - Carla
  • it-IT_CLOUD_Male - it-IT - Male - Giorgio
  • nb-NO_CLOUD_Female - no-NO - Female - Liv
  • no-NO_CLOUD_Female - no-NO - Female - Ida
  • nl-NL_CLOUD_Female - nl-NL - Female - Lotte
  • nl-NL_CLOUD_Male - nl-NL - Male - Ruben
  • pl-PL_CLOUD_Female_Agnieszka - pl-PL - Female - Agnieszka
  • pl-PL_CLOUD_Male_Jacek - pl-PL - Male - Jacek
  • pl-PL_CLOUD_Female_Ewa - pl-PL - Female - Ewa
  • pl-PL_CLOUD_Male_Jan - pl-PL - Male - Jan
  • pl-PL_CLOUD_Female - pl-PL - Female - Maja
  • pt-BR_CLOUD_Female - pt-BR - Female - Vitoria
  • pt-BR_CLOUD_Female_Camila - pt-BR - Female - Camila
  • pt-BR_CLOUD_Male - pt-BR - Male - Ricardo
  • pt-PT_CLOUD_Male - pt-PT - Male - Cristiano
  • pt-PT_CLOUD_Female - pt-PT - Female - Ines
  • ro-RO_CLOUD_Female - ro-RO - Female - Carmen
  • sv-SE_CLOUD_Female - sv-SE - Female - Astrid
  • tr-TR_CLOUD_Female - tr-TR - Female - Filiz
  • pt-BR_CLOUD_Female_Camila - pt-BR - Female - Camila

Pico TTS

  • en-US - Englisch US
  • en-GB - Englisch GB
  • de-DE - Deutsch
  • it-IT - Italiano
  • es-ES - Espaniol
  • fr-FR - Français

Coqui TTS

  • English
  • Deutsch
  • Espanol
  • Francais
  • Nederlands
  • 日本

Amazon polly direct

  • ru-RU_AP_Female - Русский - Татьяна
  • ru-RU_AP_Male - Русский - Максим
  • de-DE_AP_Female - Deutsch - Marlene
  • de-DE_AP_Female_Vicki - Deutsch - Vicki
  • de-DE_AP_Male - Deutsch - Hans
  • en-US_AP_Female - en-US - Female - Salli
  • en-US_AP_Male - en-US - Male - Joey
  • da-DK_AP_Female - da-DK - Female - Naja
  • da-DK_AP_Male - da-DK - Male - Mads
  • en-AU_AP_Female - en-AU - Female - Nicole
  • en-AU_AP_Male - en-AU - Male - Russell
  • en-GB_AP_Female_Amy - en-GB - Female - Amy
  • en-GB_AP_Male - en-GB - Male - Brian
  • en-GB_AP_Female_Emma - en-GB - Female - Emma
  • en-GB-WLS_AP_Female - en-GB-WLS - Female - Gwyneth
  • en-GB-WLS_AP_Male - en-GB-WLS - Male - Geraint
  • cy-GB_AP_Female - cy-GB - Female - Gwyneth
  • cy-GB_AP_Male - cy-GB - Male - Geraint
  • en-IN_AP_Female - en-IN - Female - Raveena
  • en-US_AP_Male_Chipmunk - en-US - Male - Chipmunk
  • en-US_AP_Male_Eric - en-US - Male - Eric
  • en-US_AP_Female_Ivy - en-US - Female - Ivy
  • en-US_AP_Female_Jennifer - en-US - Female - Jennifer
  • en-US_AP_Male_Justin - en-US - Male - Justin
  • en-US_AP_Female_Kendra - en-US - Female - Kendra
  • en-US_AP_Female_Kimberly - en-US - Female - Kimberly
  • es-ES_AP_Female - es-ES - Female - Conchita
  • es-ES_AP_Male - es-ES - Male - Enrique
  • es-US_AP_Female - es-US - Female - Penelope
  • es-US_AP_Male - es-US - Male - Miguel
  • fr-CA_AP_Female - fr-CA - Female - Chantal
  • fr-FR_AP_Female - fr-FR - Female - Celine
  • fr-FR_AP_Male - fr-FR - Male - Mathieu
  • is-IS_AP_Female - is-IS - Female - Dora
  • is-IS_AP_Male - is-IS - Male - Karl
  • it-IT_AP_Female - it-IT - Female - Carla
  • it-IT_AP_Male - it-IT - Male - Giorgio
  • nb-NO_AP_Female - nb-NO - Female - Liv
  • nl-NL_AP_Female - nl-NL - Female - Lotte
  • nl-NL_AP_Male - nl-NL - Male - Ruben
  • pl-PL_AP_Female_Agnieszka - pl-PL - Female - Agnieszka
  • pl-PL_AP_Male_Jacek - pl-PL - Male - Jacek
  • pl-PL_AP_Female_Ewa - pl-PL - Female - Ewa
  • pl-PL_AP_Male_Jan - pl-PL - Male - Jan
  • pl-PL_AP_Female - pl-PL - Female - Maja
  • pt-BR_AP_Female - pt-BR - Female - Vitoria
  • pt-BR_AP_Male - pt-BR - Male - Ricardo
  • pt-PT_AP_Male - pt-PT - Male - Cristiano
  • pt-PT_AP_Female - pt-PT - Female - Ines
  • ro-RO_AP_Female - ro-RO - Female - Carmen
  • sv-SE_AP_Female - sv-SE - Female - Astrid
  • tr-TR_AP_Female - tr-TR - Female - Filiz
  • ko-KR_AP_Female - ko-KR - Female - Seoyeon

Changelog

5.3.1 (2026-08-13)

  • (@GermanBluefox) Added the button "Test text generation" to the "Engine" tab, so the selected engine can be tested with an own text
  • (@GermanBluefox) Test generations do not overwrite the cached files anymore
  • (@GermanBluefox) Adapter requires node.js >= 22 and js-controller >= 6.0.11 now
  • (@GermanBluefox) Older changelog entries were moved to CHANGELOG_OLD.md
  • (@GermanBluefox) Updated the dependabot and auto-merge configuration

5.3.0 (2026-08-13)

  • (@GermanBluefox) Added freetts.org as a new TTS engine with more than 400 voices. An API key is required
  • (@GermanBluefox) The engine is selected in two steps now: first the provider and then the voice
  • (@GermanBluefox) The words of the Blockly block are translated into all 11 languages now
  • (@GermanBluefox) Blockly block migrated to TypeScript and the generated code is unchanged

5.2.4 (2026-08-07)

  • (@GermanBluefox) Corrected the upload of the announcement mp3 files and the location of the cache directory
  • (@GermanBluefox) Corrected the splitting of long texts for the Google engine
  • (@GermanBluefox) Corrected the detection of sonos devices in the configuration dialog
  • (@GermanBluefox) Cached files will be deleted again if the engine was changed
  • (@GermanBluefox) The queue does not block anymore if an error occurs by processing of a task
  • (@GermanBluefox) Texts with semicolons will not be interpreted as "language;volume;text" anymore
  • (@GermanBluefox) The local engines (PicoTTS, CoquiTTS) and the windows player are called without shell now
  • (@GermanBluefox) Corrected the default settings of a new instance

5.1.0 (2025-09-17)

  • (bluefox) Adapter was rewritten with TypeScript
  • (bluefox) Updated Polly voices list
  • (bluefox) Added an option to send sonos device as a parameter in sendTo command

5.0.0 (2024-07-16)

  • (mcm1957) Adapter requires admin v7 or newer now
  • (mcm1957) Adapter requires jas-controller 5 or newer now
  • (bluefox) Added possibility to play directly from states: sayit.0/tts.userfiles/gong.mp3

License

The MIT License (MIT)

Copyright (c) 2014-2026, bluefox dogafox@gmail.com

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.