Spaces:
Sleeping
Sleeping
added torch no grad
Browse files
app.py
CHANGED
@@ -32,7 +32,8 @@ def predict_and_ctc_lm_decode(input_file, model_name):
|
|
32 |
speech = load_and_fix_data(input_file)
|
33 |
|
34 |
input_values = processor(speech, return_tensors="pt", sampling_rate=16000).input_values
|
35 |
-
|
|
|
36 |
|
37 |
pred = processor.decode(logits).text
|
38 |
|
@@ -45,7 +46,8 @@ def predict_and_greedy_decode(input_file, model_name):
|
|
45 |
speech = load_and_fix_data(input_file)
|
46 |
|
47 |
input_values = processor(speech, return_tensors="pt", sampling_rate=16000).input_values
|
48 |
-
|
|
|
49 |
|
50 |
predicted_ids = torch.argmax(logits, dim=-1)
|
51 |
pred = processor.batch_decode(predicted_ids)
|
@@ -59,11 +61,11 @@ def return_all_predictions(input_file, model_name):
|
|
59 |
|
60 |
|
61 |
gr.Interface(return_all_predictions,
|
62 |
-
inputs = [gr.inputs.Audio(source="microphone", type="filepath", label="Record/ Drop audio"), gr.inputs.Dropdown(["
|
63 |
outputs = [gr.outputs.Textbox(label="Beam CTC decoding w/ LM"), gr.outputs.Textbox(label="Greedy decoding")],
|
64 |
title="ASR using Wav2Vec2 & pyctcdecode in spanish",
|
65 |
description = "Comparing greedy decoder with beam search CTC decoder, record/ drop your audio!",
|
66 |
layout = "horizontal",
|
67 |
-
examples = [["test1.wav", "
|
68 |
theme="huggingface",
|
69 |
enable_queue=True).launch()
|
|
|
32 |
speech = load_and_fix_data(input_file)
|
33 |
|
34 |
input_values = processor(speech, return_tensors="pt", sampling_rate=16000).input_values
|
35 |
+
with torch.no_grad():
|
36 |
+
logits = model(input_values).logits.cpu().detach().numpy()[0]
|
37 |
|
38 |
pred = processor.decode(logits).text
|
39 |
|
|
|
46 |
speech = load_and_fix_data(input_file)
|
47 |
|
48 |
input_values = processor(speech, return_tensors="pt", sampling_rate=16000).input_values
|
49 |
+
with torch.no_grad():
|
50 |
+
logits = model(input_values).logits
|
51 |
|
52 |
predicted_ids = torch.argmax(logits, dim=-1)
|
53 |
pred = processor.batch_decode(predicted_ids)
|
|
|
61 |
|
62 |
|
63 |
gr.Interface(return_all_predictions,
|
64 |
+
inputs = [gr.inputs.Audio(source="microphone", type="filepath", label="Record/ Drop audio"), gr.inputs.Dropdown(["LuisG07/wav2vec2-large-xlsr-53-spanish", "jonatasgrosman/wav2vec2-xls-r-1b-spanish"], label="Model Name")],
|
65 |
outputs = [gr.outputs.Textbox(label="Beam CTC decoding w/ LM"), gr.outputs.Textbox(label="Greedy decoding")],
|
66 |
title="ASR using Wav2Vec2 & pyctcdecode in spanish",
|
67 |
description = "Comparing greedy decoder with beam search CTC decoder, record/ drop your audio!",
|
68 |
layout = "horizontal",
|
69 |
+
examples = [["test1.wav", "LuisG07/wav2vec2-large-xlsr-53-spanish"], ["test2.wav", "jonatasgrosman/wav2vec2-xls-r-1b-spanish"]],
|
70 |
theme="huggingface",
|
71 |
enable_queue=True).launch()
|