Update README.md
Browse files
README.md
CHANGED
|
@@ -45,7 +45,7 @@ ds = load_dataset("common_voice", "fr", split="test", cache_dir="./data/fr")
|
|
| 45 |
|
| 46 |
|
| 47 |
|
| 48 |
-
chars_to_ignore_regex = '[
|
| 49 |
def map_to_array(batch):
|
| 50 |
speech, _ = torchaudio.load(batch["path"])
|
| 51 |
batch["speech"] = resampler.forward(speech.squeeze(0)).numpy()
|
|
@@ -79,7 +79,9 @@ print(wer.compute(predictions=result["predicted"], references=result["target"]))
|
|
| 79 |
## Testing
|
| 80 |
|
| 81 |
All the Common Voice `Test` dataset (15763 files) were used for testing.
|
|
|
|
| 82 |
Results:
|
|
|
|
| 83 |
WER=20.89%
|
| 84 |
SER=77.56%
|
| 85 |
|
| 45 |
|
| 46 |
|
| 47 |
|
| 48 |
+
chars_to_ignore_regex = '[\\,\\?\\.\\!\\-\\;\\:\\"\\β\\%\\β\\β\\οΏ½\\β\\β\\β\\β\\β\\β¦\\Β·\\!\\Η\\?\\Β«\\βΉ\\Β»\\βΊβ\\β\\\\ΚΏ\\ΚΎ\\β\\β\\\\|\\.\\,\\;\\:\\*\\β\\β\\β\\β\\_\\/\\:\\Λ\\;\\,\\=\\Β«\\Β»\\β]'
|
| 49 |
def map_to_array(batch):
|
| 50 |
speech, _ = torchaudio.load(batch["path"])
|
| 51 |
batch["speech"] = resampler.forward(speech.squeeze(0)).numpy()
|
| 79 |
## Testing
|
| 80 |
|
| 81 |
All the Common Voice `Test` dataset (15763 files) were used for testing.
|
| 82 |
+
|
| 83 |
Results:
|
| 84 |
+
|
| 85 |
WER=20.89%
|
| 86 |
SER=77.56%
|
| 87 |
|