main.py

from flask import Flask, request, jsonify, abort
import whisper
import os

from flask_cors import CORS
from tempfile import NamedTemporaryFile

# Load Whisper model
print("\nLoading Whisper\n", flush=True)
model = whisper.load_model("small")

# Initialize Flask app
app = Flask(__name__)
CORS(app)
print("\nHello, welcome to SemaBox\n", flush=True)


def transcribe(audio):
    
    #time.sleep(3)
    # load audio and pad/trim it to fit 30 seconds
    audio = whisper.load_audio(audio)
    audio = whisper.pad_or_trim(audio)

    # make log-Mel spectrogram and move to the same device as the model
    mel = whisper.log_mel_spectrogram(audio).to(model.device)

    # detect the spoken language
    _, probs = model.detect_language(mel)
    print(f"Detected language: {max(probs, key=probs.get)}")

    # decode the audio
    options = whisper.DecodingOptions(fp16 = False)
    result = whisper.decode(model, mel, options)
    return result.text
    
@app.route("/")
def hello():
    return "Semabox, listens to you!"

@app.route('/whisper', methods=['POST'])
def transcribe_audio():
    if 'audio' not in request.files:
        # If no audio file is submitted, return a 400 (Bad Request) error.
        abort(400, description="No audio file provided")

    audio_file = request.files['audio']

    # Create a temporary file to save the uploaded audio.
    with NamedTemporaryFile(suffix=".wav", delete=True) as temp:
        audio_file.save(temp.name)

        # Perform transcription using the Whisper model.
        result = model.transcribe(temp.name)

        # Return the transcribed text in JSON format.
        return jsonify({
            'filename': audio_file.filename,
            'transcript': result['text'],
        })


# Define the route for transcription
@app.route('/transcribe', methods=['POST'])
def transcribe_audio():
    # Check if an audio file is included in the request
    if 'audio' not in request.files:
        return jsonify({"error": "No audio file provided"}), 400
    
    audio_file = request.files['audio']
    
    # Save the uploaded audio file
    audio_path = os.path.join("temp_audio", audio_file.filename)
    audio_file.save(audio_path)
    
    # Transcribe the audio
    transcription, language = transcribe(audio_path)
    
    # Clean up the saved file
    os.remove(audio_path)
    
    # Return the transcription and detected language
    return jsonify({"transcription": transcription, "language": language}), 200
    
# Healthcheck endpoint
@app.route('/healthcheck', methods=['GET'])
def healthcheck():
    return jsonify({"status": "API is running"}), 200

# Run the Flask app
if __name__ == '__main__':
    app.run(host="0.0.0.0", port=5000)