In deployment, systems that use speech as input must make use of automated transcriptions. Yet, typically when these systems are evaluated, gold transcriptions are assumed. We explicitly examine the impact of transcription errors on the downstream pe
rformance of a multi-modal system on three related tasks from three datasets: emotion, sarcasm, and personality detection. We include three separate transcription tools and show that while all automated transcriptions propagate errors that substantially impact downstream performance, the open-source tools fair worse than the paid tool, though not always straightforwardly, and word error rates do not correlate well with downstream performance. We further find that the inclusion of audio features partially mitigates transcription errors, but that a naive usage of a multi-task setup does not.
