Spaces:
Sleeping
Sleeping
Update app.py
Browse files
app.py
CHANGED
@@ -13,76 +13,65 @@ model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-base-960h")
|
|
13 |
|
14 |
async def recognize_speech(websocket):
|
15 |
async for message in websocket:
|
16 |
-
|
17 |
-
|
18 |
-
|
19 |
-
|
20 |
-
|
21 |
-
|
22 |
-
|
23 |
-
|
24 |
-
# Decode predictions
|
25 |
-
predicted_ids = torch.argmax(logits, dim=-1)
|
26 |
-
transcription = tokenizer.decode(predicted_ids[0])
|
27 |
-
# Send transcription back to the client
|
28 |
-
await websocket.send(transcription)
|
29 |
-
except Exception as e:
|
30 |
-
print(f"Error in recognize_speech: {e}")
|
31 |
-
await websocket.send("Error processing audio data.")
|
32 |
|
33 |
async def main_logic():
|
34 |
async with websockets.serve(recognize_speech, "localhost", 8000):
|
35 |
await asyncio.Future() # run forever
|
36 |
|
37 |
-
#
|
38 |
-
st.title("Real-Time ASR with Transformers")
|
39 |
|
40 |
-
#
|
41 |
st.markdown("""
|
42 |
<script>
|
43 |
const handleAudio = async (stream) => {
|
44 |
const websocket = new WebSocket('ws://localhost:8000');
|
45 |
-
const mediaRecorder = new MediaRecorder(stream, {
|
46 |
const audioChunks = [];
|
47 |
|
48 |
mediaRecorder.addEventListener("dataavailable", event => {
|
|
|
49 |
audioChunks.push(event.data);
|
50 |
-
|
51 |
-
|
52 |
-
mediaRecorder.addEventListener("stop", () => {
|
53 |
-
const audioBlob = new Blob(audioChunks);
|
54 |
-
websocket.send(audioBlob);
|
55 |
});
|
56 |
|
57 |
websocket.onmessage = (event) => {
|
58 |
const transcription = event.data;
|
59 |
const transcriptionDiv = document.getElementById("transcription");
|
60 |
-
transcriptionDiv.innerHTML
|
|
|
61 |
};
|
62 |
|
|
|
|
|
63 |
websocket.onopen = () => {
|
64 |
-
console.log('
|
65 |
};
|
66 |
|
67 |
websocket.onerror = (error) => {
|
68 |
-
console.error('WebSocket
|
69 |
};
|
70 |
|
71 |
websocket.onclose = () => {
|
72 |
-
console.log('WebSocket
|
73 |
};
|
74 |
-
|
75 |
-
mediaRecorder.start(1000);
|
76 |
};
|
77 |
|
78 |
navigator.mediaDevices.getUserMedia({ audio: true })
|
79 |
.then(handleAudio)
|
80 |
-
.catch(error => console.error('Error
|
81 |
</script>
|
82 |
|
83 |
<div id="transcription">Your transcriptions will appear here:</div>
|
84 |
""", unsafe_allow_html=True)
|
85 |
|
86 |
-
# To run the WebSocket server
|
87 |
if __name__ == "__main__":
|
88 |
-
asyncio.
|
|
|
13 |
|
14 |
async def recognize_speech(websocket):
|
15 |
async for message in websocket:
|
16 |
+
wf, samplerate = sf.read(io.BytesIO(message))
|
17 |
+
input_values = tokenizer(wf, return_tensors="pt").input_values
|
18 |
+
with torch.no_grad():
|
19 |
+
logits = model(input_values).logits
|
20 |
+
|
21 |
+
predicted_ids = torch.argmax(logits, dim=-1)
|
22 |
+
transcription = tokenizer.decode(predicted_ids[0])
|
23 |
+
await websocket.send(transcription)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
24 |
|
25 |
async def main_logic():
|
26 |
async with websockets.serve(recognize_speech, "localhost", 8000):
|
27 |
await asyncio.Future() # run forever
|
28 |
|
29 |
+
# Create the streamlit interface
|
30 |
+
st.title("Real-Time ASR with Transformers.js")
|
31 |
|
32 |
+
# The script can't be run via "streamlit run" because that hangs asyncio loop
|
33 |
st.markdown("""
|
34 |
<script>
|
35 |
const handleAudio = async (stream) => {
|
36 |
const websocket = new WebSocket('ws://localhost:8000');
|
37 |
+
const mediaRecorder = new MediaRecorder(stream, {mimeType: 'audio/webm'});
|
38 |
const audioChunks = [];
|
39 |
|
40 |
mediaRecorder.addEventListener("dataavailable", event => {
|
41 |
+
console.log('dataavailable:', event.data);
|
42 |
audioChunks.push(event.data);
|
43 |
+
websocket.send(event.data);
|
|
|
|
|
|
|
|
|
44 |
});
|
45 |
|
46 |
websocket.onmessage = (event) => {
|
47 |
const transcription = event.data;
|
48 |
const transcriptionDiv = document.getElementById("transcription");
|
49 |
+
transcriptionDiv.innerHTML = transcriptionDiv.innerHTML + transcription + "<br/>";
|
50 |
+
console.log('Received:', transcription);
|
51 |
};
|
52 |
|
53 |
+
mediaRecorder.start(1000);
|
54 |
+
|
55 |
websocket.onopen = () => {
|
56 |
+
console.log('Connected to WebSocket');
|
57 |
};
|
58 |
|
59 |
websocket.onerror = (error) => {
|
60 |
+
console.error('WebSocket Error:', error);
|
61 |
};
|
62 |
|
63 |
websocket.onclose = () => {
|
64 |
+
console.log('WebSocket Closed');
|
65 |
};
|
|
|
|
|
66 |
};
|
67 |
|
68 |
navigator.mediaDevices.getUserMedia({ audio: true })
|
69 |
.then(handleAudio)
|
70 |
+
.catch(error => console.error('getUserMedia Error:', error));
|
71 |
</script>
|
72 |
|
73 |
<div id="transcription">Your transcriptions will appear here:</div>
|
74 |
""", unsafe_allow_html=True)
|
75 |
|
|
|
76 |
if __name__ == "__main__":
|
77 |
+
asyncio.run(main_logic())
|