github nimroddolev/chime_tts v0.14.0-beta1

latest releases: v1.2.3-beta1, v1.2.2, docusaurus-v1.2.2...
pre-release20 months ago

✨ New Features:

Media Players

  • Support has been added for the announce feature on unsupported media_player platforms such as HomePods and Spotify, pausing currently playing music during Chime TTS playback and restoring it afterwards.
  • Alexa media_players can now be used in service calls to chime_tts.say without additional service calls or audio conversions.
    Note: the Alexa Media Player is required, however the previous method can still be used.

TTS Audio Adjustments

  • The pitch of TTS audio can be customized via the tts_pitch parameter, allowing you to increase/decrease the pitch in semitones to add some variation and even musicality to your announcements.
  • Adjusting the speed of TTS audio has been rewritten:
    • Fewer artifacts introduced when changing the speed.
    • Support has been added for slowing down audio.
    • The speed can now be set to between 1% - 400% of the original.
    • The parameter name has been changed to tts_speed (the previous tts_playback_speed name is still supported).
  • Adjusting the volume of generated audio files has been made simpler with the addition of a new option in the audio_conversion dropdown menu.

Chimes

New chimes have been added:

  1. "Ba-Dum Tss!"
  2. "Crickets" and
  3. "Soft"
  4. "Knock"

Message Segments

  • The YAML structure for message segments has been improved to be more 50% more concise and user-friendly. The previous structure is still supported.

    Previous Structure:

      message:
        - type: chime
          path: bells_2
        - type: delay
          length: 2000
        - type: tts
          message: Someone is at the door  
    

    New Structure:

       message:
         - chime: bells_2
         - delay: 2000
         - tts: Someone is at the door
    
  • Chimes & TTS segments can now be repeated, via the new repeat attribute, eg:

      message:
        - chime: /media/sounds.cuckoo.mp3
          repeat: 3
        - tts: It's 3 o'clock
    

🧰 Improvements:

  • The "ElevenLabs TTS" platform has been added to the list of TTS platforms
  • The chime_tts.say_url response object now includes a success value as well as a media_content_id value (when the public folder in the configuration is within the media folder).
  • The folder path for audio files generated from the chime_tts.say_url service can now accept both local and public directory paths (see Chime TTS' configuration page in SettingsIntegrations).
  • A default TTS platform can now be set in the configuration and will be used if tts_platform is null
  • A default offset value can be set inthe configuration
  • When onboarding the Chime TTS integration, an info popup appears if no TTS platforms are configured, to explain that at least 1 is required.

🐞 Bugfixes:

  • Fixed bug when queueing multiple service calls.
  • Fixed bug when mp3 generation failed.
  • Downloaded chimes are now properly added to cache.
  • Fixed bug that resulted in incorrect default absolute folder paths.
  • Fixed validation for custom chime paths in configuration.
  • Fixed a typo in custom chime option "Custom 4".
  • Added support for language selection when using Microsoft Edge TTS. - fe9f0b1 jonalbr
  • Added missing Fanfare chime option to end_chime_path drop down list.
  • Downloaded chimes are now deleted if cache is set to false.
  • Fixed bug which resulted in segment-specific audio_conversion arguments being applied to the entire audio message.

What's Changed

New Contributors

Full Changelog: v0.13.0...v0.14.0-beta1

Don't miss a new chime_tts release

NewReleases is sending notifications on new releases.