Commit
•
ea52116
1
Parent(s):
425ed92
Update README.md
Browse files
README.md
CHANGED
|
@@ -7,10 +7,9 @@ Fine-tuned facebook/wav2vec2-large on English using the UISpeech Corpus. When us
|
|
| 7 |
|
| 8 |
The script used for training can be found here: https://github.com/amceejay/HIYACCENT-NE-Speech-Recognition-System
|
| 9 |
|
| 10 |
-
Usage
|
| 11 |
-
The model can be used directly (without a language model) as follows...
|
| 12 |
|
| 13 |
-
Using the ASRecognition library:
|
| 14 |
from asrecognition import ASREngine
|
| 15 |
|
| 16 |
asr = ASREngine("fr", model_path="codeceejay/HIYACCENT_Wav2Vec2")
|
|
@@ -20,7 +19,7 @@ transcriptions = asr.transcribe(audio_paths)
|
|
| 20 |
|
| 21 |
|
| 22 |
|
| 23 |
-
Writing your own inference speech:
|
| 24 |
import torch
|
| 25 |
import librosa
|
| 26 |
from datasets import load_dataset
|
|
|
|
| 7 |
|
| 8 |
The script used for training can be found here: https://github.com/amceejay/HIYACCENT-NE-Speech-Recognition-System
|
| 9 |
|
| 10 |
+
##Usage: The model can be used directly (without a language model) as follows...
|
|
|
|
| 11 |
|
| 12 |
+
#Using the ASRecognition library:
|
| 13 |
from asrecognition import ASREngine
|
| 14 |
|
| 15 |
asr = ASREngine("fr", model_path="codeceejay/HIYACCENT_Wav2Vec2")
|
|
|
|
| 19 |
|
| 20 |
|
| 21 |
|
| 22 |
+
##Writing your own inference speech:
|
| 23 |
import torch
|
| 24 |
import librosa
|
| 25 |
from datasets import load_dataset
|