Update README.md
Browse files
README.md
CHANGED
|
@@ -15,7 +15,6 @@ tags:
|
|
| 15 |
- speech
|
| 16 |
---
|
| 17 |
|
| 18 |
-
[](https://colab.research.google.com/gist/waveletdeboshir/c01334561f23c5167598b2054e50839a/gigaam-ctc-hf-finetune.ipynb)
|
| 19 |
|
| 20 |
# GigaAM-v2-CTC with ngram LM and beamsearch 🤗 Hugging Face transformers
|
| 21 |
|
|
@@ -75,7 +74,7 @@ transcription = processor.batch_decode(logits=logits.numpy()).text[0]
|
|
| 75 |
### Decoding with timestamps
|
| 76 |
We can use decoder to extract word-level timestamps. For this we need to know model stride and set parameter `output_word_offsets=True`.
|
| 77 |
|
| 78 |
-
In our case (Conformer) MODEL_STRIDE = 40 ms per timestamp
|
| 79 |
|
| 80 |
```python
|
| 81 |
MODEL_STRIDE = 40
|
|
|
|
| 15 |
- speech
|
| 16 |
---
|
| 17 |
|
|
|
|
| 18 |
|
| 19 |
# GigaAM-v2-CTC with ngram LM and beamsearch 🤗 Hugging Face transformers
|
| 20 |
|
|
|
|
| 74 |
### Decoding with timestamps
|
| 75 |
We can use decoder to extract word-level timestamps. For this we need to know model stride and set parameter `output_word_offsets=True`.
|
| 76 |
|
| 77 |
+
In our case (Conformer) `MODEL_STRIDE = 40` ms per timestamp.
|
| 78 |
|
| 79 |
```python
|
| 80 |
MODEL_STRIDE = 40
|