Automatic Speech Recognition
NeMo
Portuguese
ASR
Speech
Portuguese
parakeet
Eval Results (legacy)
teste / README.md
opedromartins's picture
Update README.md
a3caf97 verified
|
Raw History Blame Contribute Delete
2.03 kB
---
license: cc-by-4.0
library_name: nemo
base_model: nvidia/parakeet-tdt-0.6b-v3
language:
- pt
tags:
- ASR
- Speech
- Portuguese
- automatic-speech-recognition
- parakeet
datasets:
- opedromartins/ASR-datasets-ptbr
- opedromartins/ASR-datasets-ptbr-clean-transcriptions
model-index:
- name: parakeet-tdt-v3-ptBR-30s
results:
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: CORAA MuPe
type: opedromartins/ASR-datasets-ptbr-clean-transcriptions
split: test
metrics:
- type: wer
value: 10.84
name: WER
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: CORAA NURC-SP
type: opedromartins/ASR-datasets-ptbr-clean-transcriptions
split: test
metrics:
- type: wer
value: 11.77
name: WER
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: CORAA v1.1
type: opedromartins/ASR-datasets-ptbr-clean-transcriptions
split: test
metrics:
- type: wer
value: 12.52
name: WER
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: Fleurs
type: opedromartins/ASR-datasets-ptbr-clean-transcriptions
split: test
metrics:
- type: wer
value: 3.55
name: WER
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: Multilingual LibriSpeech
type: opedromartins/ASR-datasets-ptbr-clean-transcriptions
split: test
metrics:
- type: wer
value: 10.96
name: WER
- task:
type: automatic-speech-recognition
name: Automatic Speech Recognition
dataset:
name: TEDx
split: test
type: opedromartins/ASR-datasets-ptbr-clean-transcriptions
metrics:
- type: wer
value: 7.04
name: WER
---
# Parakeet-tdt-v3-ptbr-30s