Quality-Aware Self-Correcting Speech Translation on an Edge Device

arXiv:2610.07545v1 Announce Type: new Abstract: We present a fully offline speech-to-speech translation pipeline that runs on a Jetson Nano (4 GB) and corrects its own weak translations without retraining. A Whisper-tiny ASR feeds an Opus-MT translator; multilingual BERT cosine similarity acts as a Quality Estimation (QE) gate, triggering a secondary-pass correction when confidence falls below a pre-defined threshold au. We compare three correction methods: QE reranking (M1), Minimum Bayes-Risk decoding (M2), and constrained beam search (M3). On 1,012 FLORES-200 sentences (English-Spanish), M2 at au=0.90 produces statistically significant improvements over greedy decoding on BLEU (+0.67, p0.99). Our central finding is that QE functions effectively as a gate but poorly as a ranker: removing the QE model from candidate selection (M1oM2) does not hurt quality and frees 680 MB from the critical path. Using a gain-to-edit ratio adapted from the post-editing-effort literature, we further show that smaller candidate pools (N=3) yield more surgical corrections with better semantic adequacy, while larger pools (N=10) maximise lexical reward. We release the system and demonstrate live translation across six language pairs.

Related

• Automatic Labelling of Speech Translation Errors
• MetaHOPE: A Metaphor-Oriented Evaluation Framework for Analysing MT and LLM Translation Errors
• Simulstream: Open-Source Toolkit for Evaluation and Demonstration of Streaming Speech-to-Text Translation Systems

Source: arXiv cs.CL | 2026-10-07