Discriminative Axis, Not Data Volume: What a Contrastive Corpus Teaches an Audio Embedding
Scaling the corpus is the default remedy when a contrastive representation lacks an attribute. We report a case where it does nothing, and identify what does: adding a lexical-speech round to a frozen-base multimodal embedding model raises zero-shot keyword spotting by 76 points while reducing speech-emotion recognition by 14. The loss is not a capacity limit: fine-tuning on 7,442 clips from a prosody-controlled corpus recovers emotion past its pre-speech level at a five-point keyword cost. Nor is it data volume: 29,428 mined clips whose captions explicitly name emotions, at matched exposure, move emotion by -0.0007. The difference is structural: a contrastive objective encodes an attribu...
| Authors | Abdul Basit Tonmoy |
|---|---|
| Primary category | cs.CL |
| Published | 3 August 2026 |
| Last updated at source | 3 August 2026 |
Connected in the knowledge graph
- EximiusLabs/fusion-embedding-2-2b-previewModel · cited by
Models and datasets referencing this paper
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.