ksingla025
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -24,17 +24,12 @@ pip install nemo_toolkit
|
|
24 |
|
25 |
### How to run
|
26 |
|
27 |
-
|
28 |
-
import nemo.collections.asr as nemo_asr
|
29 |
|
30 |
-
|
31 |
-
|
32 |
-
asr_model = nemo_asr.models.EncDecCTCModel.from_pretrained(model_name)
|
33 |
-
|
34 |
-
# Step 2: Provide the path to your audio file
|
35 |
-
audio_file_path = '/path/to/your/audio_file.wav'
|
36 |
|
37 |
-
|
38 |
-
|
39 |
-
|
40 |
```
|
|
|
24 |
|
25 |
### How to run
|
26 |
|
27 |
+
Assuming in the Nemo docker, or have nemo_toolkit installed
|
|
|
28 |
|
29 |
+
```bash
|
30 |
+
#Make sure you change chunk_len_in_secs = 4.0,total_buffer_in_secs = 7.2, model_stride = 4
|
|
|
|
|
|
|
|
|
31 |
|
32 |
+
python NeMo/examples/asr/asr_chunked_inference/ctc/speech_to_text_buffered_infer_ctc.py \
|
33 |
+
model_path=<path-to-nemo-file> \
|
34 |
+
audio_dir=<folder-with-audio-files>
|
35 |
```
|