Transformers
Safetensors
Slovenian
wav2vec2-bert
audio-frame-classification
prosody
segmentation
audio
speech
Instructions to use classla/wav2vecbert2-prosodicUnit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use classla/wav2vecbert2-prosodicUnit with Transformers:
# Load model directly from transformers import AutoProcessor, AutoModelForAudioFrameClassification processor = AutoProcessor.from_pretrained("classla/wav2vecbert2-prosodicUnit") model = AutoModelForAudioFrameClassification.from_pretrained("classla/wav2vecbert2-prosodicUnit") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -187,7 +187,7 @@ def merge_events(events: list[list[float]], centroids):
|
|
| 187 |
# Merged list to store final intervals
|
| 188 |
merged = []
|
| 189 |
|
| 190 |
-
for event, centroid in zip(flattened_events,
|
| 191 |
if not merged:
|
| 192 |
# If merged is empty, simply add the first event
|
| 193 |
merged.append((event, centroid))
|
|
|
|
| 187 |
# Merged list to store final intervals
|
| 188 |
merged = []
|
| 189 |
|
| 190 |
+
for event, centroid in zip(flattened_events, flattened_centroids):
|
| 191 |
if not merged:
|
| 192 |
# If merged is empty, simply add the first event
|
| 193 |
merged.append((event, centroid))
|