search

SEPIA-Framework / sepia-stt-server

SEPIA server to support open-source speech recognition via WebSocket connection.
MIT License
116 stars 21 forks source link

Remove authentication requirement? #18

Open Goddard opened 1 year ago

Goddard commented 1 year ago

Just curious if some config option exists for this.

fquirin commented 1 year ago

Hi @Goddard ,

you mean like completely remove any user authentication?

[EDIT] Below comment is referring to SEPIA server NOT STT server alone:

Since everything in SEPIA is built to work based on user-IDs the system always requires some sort of account to work but in theory one could either replace the authentication module with something that simply accepts every request without checking the password or one could create a number of accounts and hard-code the login password into each client 🤔. Both options are not officially supported (yet?!).

Goddard commented 1 year ago

On local networks where the server would never be exposed to the internet the authentication isn't required at least in my use case. I was planning on having many hundred streams coming to the server or a cluster of servers.

fquirin commented 1 year ago

Oh sorry I just realized that you were talking about the STT server explicitly and not the SEPIA server in general, I mixed up the issues section 🤦‍♂️. Actually in this case you can simply stick with the default setting common_auth_token=test1234 which is already hard-coded into the STT-client (web) if no token is given. Or you can hard-code it into your custom requests.

[EDIT] The corresponding default clientId is "any".