ga642381/SpeechPrompt-v2

《SpeechPrompt v2: Prompt Tuning for Speech Classification Tasks》Speech processing with prompting paradigm

29
/ 100
Experimental

This helps researchers in machine learning or speech technology to efficiently classify spoken audio by fine-tuning pre-trained speech models. It takes raw audio recordings and their corresponding labels as input, processes them into discrete units, and then uses a prompt-tuning method to classify the audio. The output is a JSON file showing the original audio file, its true label, and the model's predicted label for each audio segment. It's designed for machine learning engineers or researchers working on speech classification tasks.

No commits in the last 6 months.

Use this if you need to perform speech classification on new audio datasets without extensive retraining of large speech models, by leveraging prompt tuning.

Not ideal if you are looking for a ready-to-use, off-the-shelf speech classification tool that doesn't require familiarity with machine learning pipeline steps.

speech-recognition audio-classification machine-learning-research natural-language-processing deep-learning-engineering
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 9 / 25
Maturity 8 / 25
Community 12 / 25

How are scores calculated?

Stars

81

Forks

8

Language

Python

License

Last pushed

Oct 19, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/prompt-engineering/ga642381/SpeechPrompt-v2"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.