sarannair/german_asr_2
收藏Hugging Face2025-11-14 更新2025-12-20 收录
下载链接：
https://hf-mirror.com/datasets/sarannair/german_asr_2
下载链接
链接失效反馈官方服务：
资源简介：
---
tags:
- audio
- asr
- speech
- text
language:
- de
license: cc-by-4.0
---
# German ASR Dataset (consolidated, with splits)
This dataset aggregates multiple German speech sources with paired transcripts.
## Splits
- **train**: ~90% of examples (shuffled)
- **validation**: ~10% of examples (shuffled)
## Fields
- **audio**: audio file (various formats)
- **text**: transcript (plain text)
## Sources
- HUI_Audio_Corpus_Clean
- Bundestag ASR (Train_Dataset)
- De_Distant_Speech_Data_Corpus_1415_Realtek
- Stadtverwaltung Boppard Meeting Recording (Youtube)
- Stadtverwaltung Koblenz Meeting Recording (Youtube)
> Upload generated automatically from on-prem GPU paths.
提供机构：
sarannair
