Update app.py
Browse files
app.py
CHANGED
|
@@ -64,13 +64,16 @@ MOS prediction demo using UTMOS-strong w/o phoneme encoder model, which is train
|
|
| 64 |
This demo only accepts .wav format. Best at 16 kHz sampling rate.
|
| 65 |
|
| 66 |
Paper is available [here](https://arxiv.org/abs/2204.02152)
|
|
|
|
|
|
|
|
|
|
| 67 |
"""
|
| 68 |
|
| 69 |
iface = gr.Interface(
|
| 70 |
fn=calc_mos,
|
| 71 |
inputs=[gr.Audio(type='filepath'), gr.Textbox(placeholder="Insert referance here", label="Referance")],
|
| 72 |
outputs=[gr.Textbox("Predicted MOS"), gr.Textbox("Hypothesis"), gr.Textbox("WER")],
|
| 73 |
-
title="
|
| 74 |
description=description,
|
| 75 |
allow_flagging="auto",
|
| 76 |
)
|
|
|
|
| 64 |
This demo only accepts .wav format. Best at 16 kHz sampling rate.
|
| 65 |
|
| 66 |
Paper is available [here](https://arxiv.org/abs/2204.02152)
|
| 67 |
+
|
| 68 |
+
Add ASR based on wav2vec-960, currently only English available.
|
| 69 |
+
Add WER interface.
|
| 70 |
"""
|
| 71 |
|
| 72 |
iface = gr.Interface(
|
| 73 |
fn=calc_mos,
|
| 74 |
inputs=[gr.Audio(type='filepath'), gr.Textbox(placeholder="Insert referance here", label="Referance")],
|
| 75 |
outputs=[gr.Textbox("Predicted MOS"), gr.Textbox("Hypothesis"), gr.Textbox("WER")],
|
| 76 |
+
title="Laronix Voice Quality Checking Demo",
|
| 77 |
description=description,
|
| 78 |
allow_flagging="auto",
|
| 79 |
)
|