Spaces:
Running
Running
HKAB
commited on
Commit
Β·
0132b7a
1
Parent(s):
ada3f10
small update
Browse files
README.md
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
---
|
2 |
-
title:
|
3 |
emoji: π»
|
4 |
colorFrom: yellow
|
5 |
colorTo: red
|
@@ -8,7 +8,7 @@ sdk_version: 5.17.1
|
|
8 |
app_file: app.py
|
9 |
pinned: false
|
10 |
license: apache-2.0
|
11 |
-
short_description: RNN-T Whisper Encoder
|
12 |
---
|
13 |
|
14 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
|
|
1 |
---
|
2 |
+
title: Vietnamese Streaming RNN-T
|
3 |
emoji: π»
|
4 |
colorFrom: yellow
|
5 |
colorTo: red
|
|
|
8 |
app_file: app.py
|
9 |
pinned: false
|
10 |
license: apache-2.0
|
11 |
+
short_description: RNN-T with Whisper Encoder
|
12 |
---
|
13 |
|
14 |
Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
|
app.py
CHANGED
@@ -101,12 +101,12 @@ def process_uploaded_file(
|
|
101 |
MyPrint(str(e))
|
102 |
return "", build_html_output(str(e), "result_item_error")
|
103 |
|
104 |
-
title = "# Streaming RNN-T
|
105 |
description = """
|
106 |
-
Visit <https://github.com/HKAB/rnnt-
|
107 |
|
108 |
-
- This model runs on CPU (Free tier) so the RTF of FP32 model is around 1.5.
|
109 |
-
- This model mights not work with your microphone since it was trained on a quite clean dataset. Try to speak loudly
|
110 |
- Although you upload a full audio file, the model will process it in a streaming fashion.
|
111 |
"""
|
112 |
|
@@ -278,17 +278,6 @@ with demo:
|
|
278 |
recorded_output = gr.Textbox(label="Recognized speech from recordings")
|
279 |
recorded_html_info = gr.HTML(label="Info")
|
280 |
|
281 |
-
gr.Examples(
|
282 |
-
examples=examples,
|
283 |
-
inputs=[
|
284 |
-
microphone,
|
285 |
-
model_type
|
286 |
-
],
|
287 |
-
outputs=[recorded_output, recorded_html_info],
|
288 |
-
fn=process_microphone,
|
289 |
-
label="Cherry-picked examples",
|
290 |
-
)
|
291 |
-
|
292 |
upload_button.click(
|
293 |
process_uploaded_file,
|
294 |
inputs=[
|
|
|
101 |
MyPrint(str(e))
|
102 |
return "", build_html_output(str(e), "result_item_error")
|
103 |
|
104 |
+
title = "# Vietnamese Streaming RNN-T"
|
105 |
description = """
|
106 |
+
Visit <https://github.com/HKAB/vietnamese-rnnt-tutorial/> for more information.
|
107 |
|
108 |
+
- This model runs on a very slow CPU (it's Free tier) so the RTF of FP32 model is around 1.5.
|
109 |
+
- This model mights not work with your microphone since it was trained on a quite clean dataset. Try to speak loudly π
|
110 |
- Although you upload a full audio file, the model will process it in a streaming fashion.
|
111 |
"""
|
112 |
|
|
|
278 |
recorded_output = gr.Textbox(label="Recognized speech from recordings")
|
279 |
recorded_html_info = gr.HTML(label="Info")
|
280 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
281 |
upload_button.click(
|
282 |
process_uploaded_file,
|
283 |
inputs=[
|