calzoneman / aeiou Goto Github PK

By uncanny coincidence, I was thinking "it would be really cool to add text-to-speech to my CyTube instance, especially with the Moonbase Alpha voice engine", Googled around for it, came upon this, and realized you made both.

I don't really want to set up a Windows VM, but your API would fit my needs perfectly. The 1024 max character limit is fine.

What kind of rate limiting do you have set up, if any? Would you be ok with a peak load of like 15 TTS requests per minute from my server? It would probably be much less than that 99.9% of the time, but that's just my estimate of a peak.

This error is issued every time I make a request to tts.cyzon.us/files/XXXXXXXXXXXXXXXXXXXXXXXXXXXXXX.wav. Do you know why?

• https://i.imgur.com/8JBNwmA.png
• https://i.imgur.com/XGFDvkd.png
• https://i.imgur.com/AUeJI4D.png

The problem is that I just made a request to the server, and got rate limited!

Header information:

• https://i.imgur.com/zfIbX9h.png

HTTP/1.1 429 Too Many Requests
Alt-Svc: h2=":443"; ma=60
Cf-Cache-Status: BYPASS
Cf-Ray: 782da36fec0bf188-CDG
Content-Type: text/html
Date: Sun, 01 Jan 2023 19:25:01 GMT
Nel: {"success_fraction":0,"report_to":"cf-nel","max_age":604800}
Report-To: {"endpoints":[{"url":"https:\/\/a.nel.cloudflare.com\/report\/v3?s=JVHa1xD36PR%2FWEHRjMZOaemN1dH3Zmeb0a4zPmsghELyBalcfQfrs3RBlV3eWO5hV3kM%2FZIsue9IcOXhBuWtAXbl6JL9REBRCaj90f6yqR8xt9NF04LpPL7W53X9XxM%3D"}],"group":"cf-nel","max_age":604800}
Content-Length: 169
Server: nghttpx

See for example #7, these questions would be better answered by providing documentation.

Can you change it to google tts?

Is there any way to setup docker container on a headless server without X server?

Can you guys share an install like or the dectalk files. I can't seem to find them online.

Hello!

I was really interested in using the API you have running (in particular, as a TTS for stream donations since it'd be really funny), but I get the No 'Access-Control-Allow-Origin' header is present on the requested resource error.

Is this intentional?

It pronounces things slightly differently than in-game and has other small differences. It also doesn't sound exactly the same as in-game. Did the game maybe apply filters over it? Is it possible you could also do it so it's a perfect replica of the in-game text to speech?

For example:
In-game aeiou
vs
Generated aeiou
and "snake? snake! snaaaaaaaaaaaaaaaaaaake!" is pronounced differently

I am using the http://tts.cyzon.us/tts?text= version.

Edit: I also forgot to mention that it get's quieter the higher pitch it is modified to be.

Any text containing [ or ] is rejected by the /tts endpoint. This started within the past hour or two, and although I'm pretty sure it's not an issue on my end, I don't have another device to test this theory.

Making a GET request to /tts with the text parameter set to either [ or %5B (URL encoded left bracket), yields the following result:

Unable to process input "["

Obviously a single [ is not a syntactically valid DecTalk instruction, but the same error occurs with a syntactically valid program like the following:

[:phoneme on]
aeiou

Unable to process input "[:phoneme on] aeiou"

@calzoneman I really appreciate the free TTS service you're running and am not expecting anything here, but I just wanted to report an issue I've frequently been running into. Often output from text inputs that are longer than a few seconds will just suddenly, randomly cut off ~5 - 10 seconds in. Sometimes trying again by adding or removing a letter totally fixes it (I assume by generating from scratch again instead of using the cut-off cached copy) and sometimes it gets to the end, or sometimes it'll cut off at a different point.

Is there anything I can do to reduce the chance of this occurring? Thank you.

Found this site here, and I really enjoy the DECTalk 'sound'. May we have the option to export the spoken phrases to .WAV?

Found your GitHub from this site here: http://tts.cyzon.us/

I want to know this due to the fact that i kind of want to use this to mess around with pitch so i can make little tunes and things. Thank you.

or is it other software?

https://github.com/dectalk/dectalk compiles on Linux now, but all the default voices sound very different to the ones on tts.cyzon.us. What version of DECtalk is running on tts.cyzon.us, and are you using extra modifiers to change things like the default pitch, voice speed, etc?