Update app.py
Browse files
app.py
CHANGED
@@ -23,6 +23,19 @@ miso_tokenizer = WhisperTokenizer.from_pretrained("mskov/whisper-small-miso", us
|
|
23 |
dataset = load_dataset("mskov/miso_test", split="test").cast_column("audio", Audio(sampling_rate=16000))
|
24 |
|
25 |
print(dataset, "and at 0[audio][array] ", dataset[0]["audio"]["array"], type(dataset[0]["audio"]["array"]), "and at audio : ", dataset[0]["audio"])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
26 |
'''
|
27 |
inputs = feature_extractor(dataset[0]["audio"]["array"], return_tensors="pt")
|
28 |
print("inputs ::: ", inputs, "and dataset type for good measure: ", type(dataset))
|
|
|
23 |
dataset = load_dataset("mskov/miso_test", split="test").cast_column("audio", Audio(sampling_rate=16000))
|
24 |
|
25 |
print(dataset, "and at 0[audio][array] ", dataset[0]["audio"]["array"], type(dataset[0]["audio"]["array"]), "and at audio : ", dataset[0]["audio"])
|
26 |
+
|
27 |
+
def transcribe(audio):
|
28 |
+
text = pipe(audio)["text"]
|
29 |
+
return text
|
30 |
+
|
31 |
+
iface = gr.Interface(
|
32 |
+
fn=transcribe,
|
33 |
+
inputs=gr.Audio(source="microphone", type="filepath"),
|
34 |
+
outputs="text",
|
35 |
+
title="Whisper Small Miso Test",
|
36 |
+
)
|
37 |
+
|
38 |
+
iface.launch()
|
39 |
'''
|
40 |
inputs = feature_extractor(dataset[0]["audio"]["array"], return_tensors="pt")
|
41 |
print("inputs ::: ", inputs, "and dataset type for good measure: ", type(dataset))
|