Cache option for audio embedded in SSML
Whenever my skill returns SSML with embedded audio, the ASK API will fetch that audio file from my server and send it to the user's Alexa.
This generates a lot of traffic on my server and might result in high latency.
It would be great if there is a caching mechanism for these audio files that prevents the ASK API to fetch the audio file on every single request.
Here are two caching strategies that might make sense:
Add an Audio Upload API to the ASK API like Facebook does for Messenger Bots: https://developers.facebook.com/docs/messenger-platform/reference/attachment-upload-api
Automatically cache fetched audio files for 2 hours or provide a ttl option in the SSML audio tag
David Karich commented
Absolutely good idea. I could also save costs on my S3 buckets.