Instructions to use TalTechNLP/espnet2_estonian with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ESPnet
How to use TalTechNLP/espnet2_estonian with ESPnet:
from espnet2.bin.asr_inference import Speech2Text model = Speech2Text.from_pretrained( "TalTechNLP/espnet2_estonian" ) speech, rate = soundfile.read("speech.wav") text, *_ = model(speech)[0] - Notebooks
- Google Colab
- Kaggle
Commit ·
a356678
1
Parent(s): 214cbd9
Update README.md
Browse files
README.md
CHANGED
|
@@ -36,6 +36,13 @@ print(text[0])
|
|
| 36 |
|
| 37 |
#### Limitations and bias
|
| 38 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
## Training data
|
| 40 |
Acoustic training data:
|
| 41 |
|
|
|
|
| 36 |
|
| 37 |
#### Limitations and bias
|
| 38 |
|
| 39 |
+
Since this model was trained on mostly broadcast speech and texts from the web, it might have problems correctly decoding the following:
|
| 40 |
+
* Speech containing technical and other domain-specific terms
|
| 41 |
+
* Children's speech
|
| 42 |
+
* Non-native speech
|
| 43 |
+
* Speech recorded under very noisy conditions or with a microphone far from the speaker
|
| 44 |
+
* Very spontaneous and overlapping speech
|
| 45 |
+
|
| 46 |
## Training data
|
| 47 |
Acoustic training data:
|
| 48 |
|