dwd
/
stt-vosk-api
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

init

master
suvodip ghosh 2025-05-31 10:15:46 +00:00
commit c398a816ae
6 changed files with 207 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
Caddyfile Normal file
View File

@ -0,0 +1,9 @@
{
admin :2019
auto_https off
}
http:// {
reverse_profile /socket.io/* localhost:8000
reverse_proxy /* localhost:8000
}

35
Dockerfile Normal file
View File

@ -0,0 +1,35 @@
FROM python:3.9-slim as base
# Install dependencies
RUN apt-get update && apt-get install -y \
python3-pip \
ffmpeg \
wget \
&& rm -rf /var/lib/apt/lists/*
# Install Python requirements
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
# Download Vosk model
RUN wget https://alphacephei.com/vosk/models/vosk-model-small-en-us-0.15.zip && \
unzip vosk-model-small-en-us-0.15.zip && \
mv vosk-model-small-en-us-0.15 model && \
rm vosk-model-small-en-us-0.15.zip
# Copy application code
COPY app.py .
COPY Caddyfile .
FROM base as production
# Install gunicorn and eventlet
RUN pip install gunicorn eventlet
# Expose ports (8000 for app, 2019 for Caddy admin)
EXPOSE 8000 2019
CMD ["caddy", "run", "--config", "/app/Caddyfile"]
FROM base as development
# For development with auto-reload
CMD ["python", "app.py"]

37
app.py Normal file
View File

@ -0,0 +1,37 @@
import eventlet
eventlet.monkey_patch()
from flask import Flask, send_from_directory
from flask_socketio import SocketIO
from vosk import Model, KaldiRecognizer
import json
import os
from dotenv import load_dotenv
load_dotenv()
app = Flask(__name__, static_folder='static')
socketio = SocketIO(app, cors_allowed_origins="*", async_mode='eventlet')
# Load the model
model = Model("model")
@app.route('/')
def serve_index():
return send_from_directory('static', 'index.html')
@socketio.on('audio_stream')
def handle_audio_stream(audio_data):
recognizer = KaldiRecognizer(model, 16000)
recognizer.SetWords(True)
if recognizer.AcceptWaveform(audio_data):
result = json.loads(recognizer.Result())
socketio.emit('transcription', result.get('text', ''))
else:
partial = json.loads(recognizer.PartialResult())
socketio.emit('partial_transcription', partial.get('partial', ''))
if __name__ == '__main__':
os.makedirs('static', exist_ok=True)
socketio.run(app, host='0.0.0.0', port=8000)

17
docker-compose.yml Normal file
View File

@ -0,0 +1,17 @@
version: '3.8'
services:
app:
build:
context: .
target: production
ports:
- "80:80"
- "443:443"
- "2019:2019"
volumes:
- ./static:/app/static
- ./Caddyfile:/app/Caddyfile
environment:
- FLASK_ENV=production
restart: unless-stopped

6
requirements.txt Normal file
View File

@ -0,0 +1,6 @@
vosk>=0.3.44
flask>=2.0.0
flask-socketio>=5.0.0
eventlet>=0.33.0
python-dotenv>=0.19.0
caddy>=2.4.6

103
static/index.html Normal file
View File

@ -0,0 +1,103 @@
<!DOCTYPE html>
<html>
<head>
<title>Vosk Audio Streaming</title>
<script src="https://cdn.socket.io/4.5.4/socket.io.min.js"></script>
<style>
body { font-family: Arial, sans-serif; max-width: 800px; margin: 0 auto; padding: 20px; }
#transcript { border: 1px solid #ccc; padding: 15px; min-height: 100px; margin-top: 20px; }
button { padding: 10px 15px; background: #007bff; color: white; border: none; cursor: pointer; }
button:disabled { background: #cccccc; }
</style>
</head>
<body>
<h1>Real-time Speech Recognition</h1>
<button id="startBtn">Start Listening</button>
<button id="stopBtn" disabled>Stop</button>
<div id="transcript"></div>
<script>
const socket = io('http://localhost:5000');
const startBtn = document.getElementById('startBtn');
const stopBtn = document.getElementById('stopBtn');
const transcriptDiv = document.getElementById('transcript');
let mediaStream;
let audioContext;
let processor;
let microphone;
// Handle server responses
socket.on('transcription', (text) => {
transcriptDiv.innerHTML += `<p>${text}</p>`;
});
socket.on('partial_transcription', (text) => {
const lastP = transcriptDiv.lastElementChild;
if (lastP && lastP.classList.contains('partial')) {
lastP.textContent = text;
} else {
const p = document.createElement('p');
p.className = 'partial';
p.textContent = text;
transcriptDiv.appendChild(p);
}
});
// Start recording
startBtn.addEventListener('click', async () => {
try {
startBtn.disabled = true;
stopBtn.disabled = false;
transcriptDiv.innerHTML = '';
mediaStream = await navigator.mediaDevices.getUserMedia({ audio: true });
audioContext = new (window.AudioContext || window.webkitAudioContext)();
microphone = audioContext.createMediaStreamSource(mediaStream);
processor = audioContext.createScriptProcessor(4096, 1, 1);
processor.onaudioprocess = (e) => {
const audioData = e.inputBuffer.getChannelData(0);
const raw = convertFloat32ToInt16(audioData);
socket.emit('audio_stream', raw);
};
microphone.connect(processor);
processor.connect(audioContext.destination);
} catch (error) {
console.error('Error:', error);
alert('Error accessing microphone: ' + error.message);
resetControls();
}
});
// Stop recording
stopBtn.addEventListener('click', () => {
if (mediaStream) {
mediaStream.getTracks().forEach(track => track.stop());
}
if (microphone && processor) {
microphone.disconnect();
processor.disconnect();
}
resetControls();
});
// Helper functions
function resetControls() {
startBtn.disabled = false;
stopBtn.disabled = true;
}
function convertFloat32ToInt16(buffer) {
const l = buffer.length;
const buf = new Int16Array(l);
for (let i = 0; i < l; i++) {
buf[i] = Math.min(1, buffer[i]) * 0x7FFF;
}
return buf.buffer;
}
</script>
</body>
</html>