Rendering of transported audio data

The invention provides determining a loss of a packet comprising first and second audio data. As the audio data are transported in the same packet for contemporaneous rendering, they suffer the same transportation loss. The effect of the loss is remedied by generating substitution audio data differently for the first audio data and second audio data. The generating is configured to conceal (obscure) the packet loss by producing audio data having characteristics as expected which creates perceptual similarity. The first generation process can, for example, use silence or simulated noise to mimic speech, without attempting to reproduce previous speech. The second generation process can, for example, mimic background noise and reproduce long-term characteristics of previous background noise. In examples, the packet further comprises metadata for positioning the audio source in three dimensions. The invention may provide packet loss concealment for Immersive Voice and Audio Services (IVAS) calls, for example, in mobile radio telecommunication devices or in Metadata-assisted spatial audio (MASA) format optimized for direct mobile device use.