-
Notifications
You must be signed in to change notification settings - Fork 13
No custom Wakeword can be created #108
Comments
I'm having the exact same problem, got Wake Word working in my React Native app and wanted to now create my own wake word. |
Having the same issue, tried emailing them and have heard no reply. |
Hello, @abury I am facing the same problem I also send them an email but no response. I found this project https://github.com/mpc6/WakeWord-Detection I was able to make it work and generate tflite models but when I added them to my app they do not work |
@jonra1993 I tried the tflite model files from the WakeWord-Detection project you linked, replacing the sample "SpokeStack" wake word models which were working in my app, and also had no luck, they didn't seem compatible, Java Buffer errors for the CRNN models and a different error fo the Wavenet models. I'm also definitely interested in finding a solution. |
Hello, @SeanAtHealthJay I am also trying to find a solution WakeWord-Detection did not work for me either but I think it can be a good starting point. |
Hi everyone, as you have surmised, Spokestack is no longer accepting new customers. Feel free to examine our open-source SDKs at https://github.com/spokestack to gather ideas for how to solve your problems—using the documentation at https://spokestack.io/docs, it’s possible to even create models yourself! Also happy to have you continue the discussion of other approaches to solving your problems here of course. |
Describe the bug
I am trying to create a custom wake word using spokestack website but when I try to upgrade my account I get this message.
According to this blog, it seems that no new customers are allowed https://forum.spokestack.io/t/unable-to-pay-stripe-error/212
Expected behavior
Activate maker plan again or include on documentation the correct way to train a custom wakeword model using Tensor flow and jupyter.
Actual behavior
No custom wake word can be created
The text was updated successfully, but these errors were encountered: