This post is short and simple: your mp3 or ogg files played on VoxEngine scenario level with call.startPlayback or using Player will be played on the Web or Mobile SDK side in HD quality (48KHz), or on SIP side if it does support wideband audio codecs (Speex or Opus). It also appeared that Opus has 3 encoding presets - auto / speech / music, currently we use auto, but maybe we will let developers decide which preset can be used on VoxEngine scenario level later.

Сompliant speech-to-speech voice AI with Realtime API
Use Realtime API in VoxEngine to build speech-to-speech voice AI with data residency control. Set baseUrl to any OpenAI Realtime-compatible endpoint, including Azure EU deployments, and control where call audio is processed



