assemblyai / assemblyai-go-sdk Goto Github PK

Ideally I would be able to make a request to get the temporary token https://www.assemblyai.com/docs/guides/real-time-streaming-transcription#creating-temporary-authentication-tokens

The structs are there, but are not used anywhere. Either the package should be updated to include that function, or there should be a generic http function to make calls to unimplemented AssemblyAI endpoints without having to do all of the set up manually for one endpoint

https://github.com/AssemblyAI/assemblyai-go-sdk/blob/main/types.go#L90
https://github.com/AssemblyAI/assemblyai-go-sdk/blob/main/types.go#L280

Hello and thanks for the recent release, especially the features around ForceEndUtterance and SetEndUtteranceSilenceThreshold. Will be using those updates in the future.

I had a question about how the realtime portion of this SDK formats and sends audio. I noticed on the official API docs it says:

When sending audio over the WebSocket connection, you can use the websocket's binary mode to send raw audio data. This can be the raw data recorded directly from a microphone or read from an audio file.

and slightly lower down...

HEADS UP
Sending audio_data via JSON is also supported but will be deprecated in the future. Use the binary mode instead.

If I look in the codebase I find the following:

Given that &msg here is using a json.RawMessage type, it seems the realtime SDK is sending the bytes of a json message, but not the raw audio bytes themselves. Is that coming in a future release?

Its possible I am mistaken in how the SDK is working or I don't understand what the API docs are referring to when it comes to "raw audio bytes"? I would assume it means PCM bytes. Looking for clarification.

Thanks for the help!