cache: and cache_dir: seem to speak for themselves, but where is this documented? I also found tts scripts with the cache: false setting added, indicating that specific script shouldn’t be cached?
I didn’t have any of these settings, and only by coincidence found about 1200 recorded messages in my tts folder…
what is the advantage of the cache, is it to speedup frequently used (I take it fixed) messages? No use caching messages that change all the time I would think?
am I correct assuming that’s the global setting for HA, and don’t need extra settings on cache per script using the tts.google_say service? Unless Id like to do cache a dedicated script?
then, there’s the time_memory: I found. What does that do?
the language: doesn’t really seem to matter, because I can add any language to the scripts, without having them in the configuration at all…
I agree… One of the most frustrating things about the HA documentation is where it is so incomplete that it doesn’t even mention all the config options, let alone document what they do and how it works.
I don’t even know how you found those extra options but I too am interested in answers to some of the questions you ask!!
As an aside, I long ago discovered that my tts folder was filling up so I wrote a nightly automation to clear it. There is even a dedicated service to do this…
#=======================
#== Clear the TTS cache
#=======================
- alias: System Clear TTS Cache
trigger:
platform: time
at: '03:00:00'
action:
service: tts.clear_cache
The link to this doc from the actual Google Translate Text-to-Speech page is not exactly well signposted. Although now I know those links on the right panel might add relevant information I’ll remember to look in future
tbh, I still don’t see it in the right panel…where did you find this? entering tts in the integrations page reveals a sub_set of the available integrations …(google, microsoft etc aren’t listed):
the page @nickrout lists (thanks! exactly what I was searching for) is nowhere to be found, coming from either of these integrations?
anyway, having said that, the doc now at least confirms my thoughts about the language: variable: It isn’t there. While it is listed under google_translate…
So there are global tts variables and per integration variables, which makes sense I guess…
still, this is a bit confusing:
The following optional parameters can be used with any platform. However, the TTS component will only look for global settings under the configuration of the first configured platform:
what is the first platform… I have them configured in packages. Also, does this mean we can set ‘global’ variables on the ‘first’ platform, and assume the same settings are used for the others if no variables are set. Which would imply we could set variables with different values if so desired, or the dedicated platform variables (in this case the language setting which is valid for Google TTS) or the many configurable ones for eg Microsoft tts, which is what I was looking for. (really don’t like the single female voice Google offers… especially in my native language…
all in all, this is not as clear as it could be… needs further experimenting, which of course is always fun to do
And let me know how you get on with Microsoft TTS. I tried a while back and crashed my system!!! I wasn’t the only one, when I posted here to ask if anyone used it at least one other person had the same problem as me. It was a long time ago though…
no more caching, as I don’t see why it should in the first place (not explained, and all my messages are created on the fly in the first place. caching only takes disk space and causes read/write operations, which as I understand it take life form your SD’s…)
I can select any language in my setup, and it starts speaking funny
Nice. I’m going to play with the language setting too.
I’m in the process of re-writing my announcement package, and my settings page is bigger than yours
