[Feature Request] Ability to use a script for TTS.

[Chaotic]

I'd love if I could have the TTS function call the sonos_says script so that it would resume any music that happened to be playing on the sonos.

I'm not sure how challenging this change would be through.

[kalkih]

Hmm, I'm not sure how well that would work, seems like you need to specify a delay (basically how long the tts message is and when to resume playback, I guess?). This works ok for predefined messages.

Not sure how we would calculate the delay, and if we set a predefined delay, messages could get cut off, etc.

[Chaotic]

It looks like there is a way to have the script calc the delay but I haven't tested that out yet.

https://community.home-assistant.io/t/sonos-tts-script/8896/21

I'll play around with it and see if I can get it to actually work and report back.

[kalkih]

Cool, thanks, should be easy to implement if the delay is calculated correctly.

[Chaotic]

Yeah that code seems to calculate the delay correctly

sonos_say:
  alias: "Sonos TTS script"
  sequence:
    - service: media_player.sonos_snapshot
      data_template:
        entity_id: "{{ 'media_player.' ~ sonos_entity }}"
    - service: media_player.sonos_unjoin
      data_template:
        entity_id:  "{{ 'media_player.' ~ sonos_entity }}"
    - service: media_player.volume_set
      data_template:
        entity_id:  "{{ 'media_player.' ~ sonos_entity }}"
        volume_level: "{{ volume }}"
    - service: tts.google_say
      data_template:
        entity_id:  "{{ 'media_player.' ~ sonos_entity }}"
        message: "{{ message }}"
    - delay: '00:00:01'
    - delay: >-
        {% set duration = states.media_player[sonos_entity].attributes.media_duration %}
        {% if duration > 0 %}
          {% set duration = duration - 1 %}
        {% endif %}
        {% set seconds = duration % 60 %}
        {% set minutes = (duration / 60)|int % 60 %}
        {% set hours = (duration / 3600)|int %}
        {{ [hours, minutes, seconds]|join(':') }}
    - service: media_player.sonos_restore
      data_template:
        entity_id:  "{{ 'media_player.' ~ sonos_entity }}"

That's the code I ended up using only difference from the one in the link is it takes a volume input too.

I'd think the TTS from your card could just pass along whatever volume the player is set to.

[kalkih]

Now available with the latest release, wasn't able to actually test it since I don't own any sonos devices. Let me now if you have any issues.

# example
- type: custom:mini-media-player
  entity: media_player.example
  tts:
    platform: sonos
    volume: 0.75 # volume level for tts

[sandman32]

FYI the script listed here wasn't working for me, it wouldn't resume after the TTS message was delivered. However this script worked. I got it from arsaboo here: https://github.com/arsaboo/homeassistant-config/blob/master/scripts.yaml . I then modified it to list one of my Sonos devices instead of the device he listed, and I increased the first delay from 00:00:00 to 00:00:01. 00:00:00 was cutting off my tts message.

    alias: "Sonos TTS script"
    sequence:
      - service: media_player.sonos_snapshot
        data_template:
          entity_id: "{{ sonos_entity|default('media_player.family_room') }}"
      - service: media_player.sonos_unjoin
        data_template:
          entity_id: "{{ sonos_entity|default('media_player.family_room') }}"
      - service: media_player.volume_set
        data_template:
          entity_id: "{{ sonos_entity|default('media_player.family_room') }}"
          volume_level: "{{ volume|default(0.5) }}"
      - service: tts.google_say
        data_template:
          entity_id: "{{ sonos_entity|default('media_player.family_room') }}"
          message: "{{ message }}"
      - delay: "{{ delay|default('00:00:01') }}"
      - wait_template: "{{ is_state(sonos_entity|default('media_player.family_room'), 'playing') }}"
        timeout: '00:00:05'
      - service: media_player.sonos_restore
        data_template:
          entity_id: "{{ sonos_entity|default('media_player.family_room') }}"     ```

[sandman32]

Sorry that was actually cutting off my tts message if it was longer than a few seconds. This seems to be the correct answer.

sonos_say:
    alias: "Sonos TTS script"
    sequence:
      - service: media_player.sonos_snapshot
        data_template:
          entity_id: "{{ sonos_entity|default('media_player.family_room') }}"
      - service: media_player.sonos_unjoin
        data_template:
          entity_id: "{{ sonos_entity|default('media_player.family_room') }}"
      - service: media_player.volume_set
        data_template:
          entity_id: "{{ sonos_entity|default('media_player.family_room') }}"
          volume_level: "{{ volume|default(0.5) }}"
      - service: tts.google_say
        data_template:
          entity_id: "{{ sonos_entity|default('media_player.family_room') }}"
          message: "{{ message }}"
      - delay: "{{ delay|default('00:00:01') }}"
      - wait_template: "{{ is_state(sonos_entity|default('media_player.family_room'), 'playing') }}"
        timeout: '00:00:03'
      - wait_template: "{{ not is_state(sonos_entity|default('media_player.family_room'), 'playing') }}"
        timeout: '00:01:00'        
      - service: media_player.sonos_restore
        data_template:
          entity_id: "{{ sonos_entity|default('media_player.family_room') }}"

[Royal2000H]

Sorry to revive such an old topic. This is a new solution to help future readers because of the breaking change in HA 0.93 that changed sonos services to the sonos domain (Ex: media_player.sonos_snapshot to sonos.snapshot)

In regards to the above, @Chaotic's script actually worked just fine for me after some tweaks. @sandman32's works fine too after tweaks. They work similarly - the former waits a second, finds the duration of the currently playing item (the tts), sets the delay to that length, waits the delay, then resumes audio. The latter waits a second, makes sure something (the tts) is playing, then waits until that something (the tts) stops playing - upto 1 minute, then resumes audio.
In other words, they're very similar. They have similar performance in my limited experience testing both. @Chaotic's might be more flexible because of the lack of the 1 minute timeout, though that too can be adjusted. But if it doesn't work for you, you can use @sandman32's.

Note: Neither will resume playback of music you play through Spotify app onto your Sonos speaker (Spotify Connect) because the Sonos API has no way that I can find to restart that.

Here are the edited versions of both that have both been tested with HA 0.107.7 and Mini Media Player 1.7.0.

@Chaotic's version (I fixed the domains and also fixed the sonos_entity to assume "media_player." is being passed in the entity name). I chose to use this one.

sonos_say:
    alias: "Sonos TTS script"
    sequence:
    - service: sonos.snapshot
      data_template:
        entity_id: "{{ sonos_entity }}"
    - service: sonos.unjoin
      data_template:
        entity_id:  "{{ sonos_entity }}"
    - service: media_player.volume_set
      data_template:
        entity_id:  "{{ sonos_entity }}"
        volume_level: "{{ volume }}"
    - service: tts.google_say
      data_template:
        entity_id:  "{{ sonos_entity }}"
        message: "{{ message }}"
    - delay: '00:00:01'
    - delay: >-
        {% set duration = states[sonos_entity].attributes.media_duration %}
        {% if duration > 0 %}
          {% set duration = duration - 1 %}
        {% endif %}
        {% set seconds = duration % 60 %}
        {% set minutes = (duration / 60)|int % 60 %}
        {% set hours = (duration / 3600)|int %}
        {{ [hours, minutes, seconds]|join(':') }}
    - service: sonos.restore
      data_template:
        entity_id:  "{{ sonos_entity }}"

@sandman32's version (I fixed the domains and also removed the defaults referencing a random speaker because they are irrelevant)

sonos_say:
    alias: "Sonos TTS script"
    sequence:
      - service: sonos.snapshot
        data_template:
          entity_id: "{{ sonos_entity }}"
      - service: sonos.unjoin
        data_template:
          entity_id: "{{ sonos_entity }}"
      - service: media_player.volume_set
        data_template:
          entity_id: "{{ sonos_entity }}"
          volume_level: "{{ volume|default(0.5) }}"
      - service: tts.google_say
        data_template:
          entity_id: "{{ sonos_entity }}"
          message: "{{ message }}"
      - delay: "{{ delay|default('00:00:01') }}"
      - wait_template: "{{ is_state(sonos_entity, 'playing') }}"
        timeout: '00:00:03'
      - wait_template: "{{ not is_state(sonos_entity, 'playing') }}"
        timeout: '00:01:00'        
      - service: sonos.restore
        data_template:
          entity_id: "{{ sonos_entity }}"