Where is the Annyang lib?

[toddpi314]

This node module looks like a wrapper around the Annyang library, but the library is apparently not present in your repo.

Did you forget a checkin? If not, could you share how you are running the lib in node process to have the module work as expected?

[alanjames1987]

This is not a wrapper around annyang. As the README.md says "This is partial port of the amazing annyang client side library".

The thing that has been ported was the phrase routing, not the speech recognition itself.

Annyany is dependent on Chrome's speech recognition system, which obviously has not been ported to Node.

[alanjames1987]

If you would like to help support speech recognition please do. Contribution is always welcome.

[zedd45]

I had a similar expectation, due to the name under which this module was published (annyang). Unfortunately, it's the same name as the module on which it is based.

I installed this expecting my project (which is using webpack) to pick up the client side library.

[alanjames1987]

If you think I should republish this module as annyang-server or something like that I will do so. Give me your suggestions.

Additionally, if you want to help me add the features you are looking for I will accept pull requests and add any features I can add.

[zedd45]

I personally believe that annyang-node would be a great name, as it matches your repo name, but that's just my <2Cents> :)

I would love to work on the speech recog. bit, but I am not sure I have the time presently to invest in that, if it's not something we can get from a library. I'm not sure where it's built into webkit, but probably worth a look...

[alanjames1987]

annyang-node would not be a great name according to NPM's naming recommendations.

Don't put "js" or "node" in the name.

https://docs.npmjs.com/files/package.json

But a name that denotes that this only works on Node would be okay with me, which is why I suggested annyang-server.

The speech recognition aspect of Chrome streams flac files to Google servers for transcription creation. The actual speech recognition is not built into Chrome. This is what client side annyang uses. We can transcode and send those same files over to Google servers with Node, but a flac converter would have to be installed on the machine using this library.

[zedd45]

Thanks for the clarification(s) @alanjames1987. I think annyang-server is a good name.

Transcoding would be a fun project. I would like to look into this, but I don't have bandwidth for that undertaking, right now.