comodoro commited on
Commit
07d71a1
1 Parent(s): 227ae34

Simplest speech to text example

Browse files
Files changed (4) hide show
  1. .gitignore +1 -0
  2. app.py +28 -0
  3. packages.txt +1 -0
  4. requirements.txt +2 -0
.gitignore ADDED
@@ -0,0 +1 @@
 
 
1
+ /venv
app.py ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from transformers import pipeline, AutoFeatureExtractor, AutoTokenizer, Wav2Vec2ForCTC
2
+ import gradio as gr
3
+ import time
4
+
5
+ model_id = 'comodoro/wav2vec2-xls-r-300m-cs-250'
6
+ feature_extractor = AutoFeatureExtractor.from_pretrained(model_id)
7
+ model = Wav2Vec2ForCTC.from_pretrained(model_id)
8
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
9
+
10
+ p = pipeline("automatic-speech-recognition", chunk_length_s=5, model=model,
11
+ tokenizer=tokenizer, feature_extractor=feature_extractor)
12
+
13
+ def transcribe(audio, state=""):
14
+ time.sleep(2)
15
+ text = p(audio)["text"]
16
+ state += text + " "
17
+ return state
18
+
19
+ with gr.Blocks() as blocks:
20
+ audio = gr.Audio(source="microphone", type="filepath",
21
+ label='Pokud je to třeba, povolte mikrofon pro tuto stránku, \
22
+ klikněte na Record from microphone, po dokončení nahrávání na Stop recording a poté na Rozpoznat')
23
+ btn = gr.Button('Rozpoznat')
24
+ output = gr.Textbox(show_label=False)
25
+ btn.click(fn=transcribe, inputs=[audio,],
26
+ outputs=[output,])
27
+
28
+ blocks.launch(enable_queue=True, debug=True)
packages.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ ffmpeg
requirements.txt ADDED
@@ -0,0 +1,2 @@
 
 
 
1
+ torch
2
+ transformers