Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# Indo-Aryan Speech Tagger - Conformer CTC Model
|
| 2 |
|
| 3 |
This speech tagger performs transcription for 5 Indian Languages: Hindi, Punjabi, Marathi, Bengali and Gujarati. It annotates key entities, predicts speaker age, dialect and intent.
|
|
@@ -32,4 +56,4 @@ print(transcription[0])
|
|
| 32 |
## License & Attribution
|
| 33 |
|
| 34 |
Please cite AI4Bharat when using this model:
|
| 35 |
-
https://indicvoices.ai4bharat.org/
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: cc-by-4.0
|
| 3 |
+
language:
|
| 4 |
+
- bn
|
| 5 |
+
- hi
|
| 6 |
+
- pa
|
| 7 |
+
- mr
|
| 8 |
+
- gu
|
| 9 |
+
base_model:
|
| 10 |
+
- parthiv11/stt_hi_conformer_ctc_large_v2
|
| 11 |
+
tags:
|
| 12 |
+
- speech_recognition
|
| 13 |
+
- entity_tagging
|
| 14 |
+
- dialect_prediction
|
| 15 |
+
- gender
|
| 16 |
+
- age
|
| 17 |
+
- intent
|
| 18 |
+
library_name: nemo
|
| 19 |
+
datasets:
|
| 20 |
+
- WhissleAI/indicvoices_hi_tagged_transcripts
|
| 21 |
+
- WhissleAI/indicvoices_pa_tagged_transcripts
|
| 22 |
+
- WhissleAI/indicvoices_mr_tagged_transcripts
|
| 23 |
+
---
|
| 24 |
+
|
| 25 |
# Indo-Aryan Speech Tagger - Conformer CTC Model
|
| 26 |
|
| 27 |
This speech tagger performs transcription for 5 Indian Languages: Hindi, Punjabi, Marathi, Bengali and Gujarati. It annotates key entities, predicts speaker age, dialect and intent.
|
|
|
|
| 56 |
## License & Attribution
|
| 57 |
|
| 58 |
Please cite AI4Bharat when using this model:
|
| 59 |
+
https://indicvoices.ai4bharat.org/
|