WenetSpeech-Wu
Collection
4 items
•
Updated
•
3
task
stringclasses 5
values | utt_id
stringlengths 23
39
| audio
unknown | label
stringlengths 1
169
|
|---|---|---|---|
age
|
219299855797_qmvti_126_12560_seg00
| "UklGRiT1BQBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAAZGF0YQD1BQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA(...TRUNCATED)
|
青年
|
age
|
219299855797_qmvti_139_18220_seg00
| "UklGRiSVBgBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAAZGF0YQCVBgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA(...TRUNCATED)
|
青年
|
age
|
219299855859_wQUOq_20_6360_seg00
| "UklGRmT0AgBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAAZGF0YUD0AgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA(...TRUNCATED)
|
老年
|
age
|
219299855859_wQUOq_22_7560_seg00
| "UklGRiTrAABXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAAZGF0YQDrAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA(...TRUNCATED)
|
老年
|
age
|
219299855859_wQUOq_22_7560_seg01
| "UklGRiSWAABXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAAZGF0YQCWAAAIAP7/9/8HAA8A+//0//z/DAAEAAAADwAJAAYA+P/(...TRUNCATED)
|
老年
|
age
|
219299855859_wQUOq_22_7560_seg02
| "UklGRqS1AQBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAAZGF0YYC1AQAZAA0AAQD9/wAAFAALAAgAFAAdABMAFAAZAAoAFwA(...TRUNCATED)
|
老年
|
age
|
219299855859_wQUOq_230_5960_seg00
| "UklGRiStAgBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAAZGF0YQCtAgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA(...TRUNCATED)
|
老年
|
age
|
219299855859_wQUOq_231_10680_seg00
| "UklGRuQNBQBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAAZGF0YcANBQAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA(...TRUNCATED)
|
老年
|
age
|
219299855859_wQUOq_339_5720_seg01
| "UklGRqQVAQBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAAZGF0YYAVAQBxACAAv/92/4P/uf9kAPcAHQG5AAUAyP+1/6L/5/+(...TRUNCATED)
|
老年
|
age
|
219299855862_Y9EIQ_342_6240_seg00
| "UklGRiTQAgBXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAAZGF0YQDQAgAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA(...TRUNCATED)
|
老年
|
End of preview. Expand
in Data Studio
WenetSpeech-Wu Bench
We introduce WenetSpeech-Wu-Bench, the first publicly available, manually curated benchmark for Wu dialect speech processing, covering ASR, Wu-to-Mandarin AST, speaker attributes, emotion recognition, TTS, and instruct TTS, and providing a unified platform for fair evaluation.
- ASR: Wu dialect ASR (9.75 hour, including Shanghainese, Suzhounese, and Mandarin code-mixed speech). Evaluated by CER.
- Wu→Mandarin AST: Speech translation from Wu dialects to Mandarin (3k utterances, 4.4h). Evaluated by BLEU.
- Speaker Attributes & Emotion: Speaker gender/age prediction and emotion recognition on Wu dialect. Evaluated by classification accuracy.
- TTS: Wu dialect TTS with speaker prompting (242 sentences, 12 speakers). Evaluated by speaker similarity, CER, and MOS.
- Instruct TTS: Instruction-following TTS with prosodic and emotional control. Evaluated by automatic accuracy and subjective MOS.
- Downloads last month
- 219
Size of downloaded dataset files:
2.38 GB
Size of the auto-converted Parquet files:
2.38 GB
Number of rows:
13,351