REVIEW 2 cited by
Prompting the Hidden Talent of Web-Scale Speech Models for Zero-Shot Task Generalization
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
We investigate the emergent abilities of the recently proposed web-scale speech model Whisper, by adapting it to unseen tasks with prompt engineering. We selected three tasks: audio-visual speech recognition (AVSR), code-switched speech recognition (CS-ASR), and speech translation (ST) on unseen language pairs. We design task-specific prompts, by either leveraging another large-scale model, or simply manipulating the special tokens in the default prompts. Experiments show that compared to the default prompts, our proposed prompts improve performance by 10% to 45% on the three zero-shot tasks, and even outperform SotA supervised models on some datasets. In addition, our experiments reveal many interesting properties of Whisper, including its robustness to prompts, bias on accents, and the multilingual understanding in its latent space. Code is available at https://github.com/jasonppy/PromptingWhisper
Forward citations
Cited by 2 Pith papers
-
Refining Transcripts With TV Subtitles by Prompt-Based Weakly Supervised Training of ASR
Using subtitles as prompts and pseudo transcripts as targets, with a Gini-based attention weighting, refines Whisper's transcripts on low-resource Flemish TV speech without verbatim labels.
-
Optimizing ASR for Catalan-Spanish Code-Switching: A Comparative Analysis of Methodologies
For Catalan-Spanish code-switching ASR, fine-tuning Whisper on 17 hours of synthetic TTS data and decoding with the Catalan token outperforms audio concatenation and in-domain fine-tuning on held-out parliamentary test sets.
Discussion (0). Continue with ORCID to comment.