Dataset · k9cli
Data shard from video-to-audio slicing pipeline for ASR and TTS
k9cli/video-vec2wav2-tokenizer-3
Continuation of the speech recognition and synthesis dataset export gathered from video files.
What's inside
This is the third shard part of the project converting video into training datasets for ASR and TTS tasks. Judging by the pipeline description, source videos are converted to 16 kHz PCM mono audio, transcribed via Whisper, and sliced into clips. Exact volume and languages are not specified in the card.
What you can do with it
- Training and fine-tuning automatic speech recognition (ASR) models
- Preparing data for text-to-speech (TTS) synthesis models
- Downloads this month
- 433K
- Likes
- 14
How to load it
from datasets import load_dataset
ds = load_dataset("k9cli/video-vec2wav2-tokenizer-3")