stt-whisper-cpp-ap/app.py

62 lines
1.7 KiB
Python

from flask import Flask, request, jsonify
from flask_cors import CORS
import os
import tempfile
import subprocess
app = Flask(__name__)
CORS(app)
WHISPER_CPP_PATH = "/whisper.cpp"
MODEL_PATH = "/whisper.cpp/models/ggml-base.en.bin"
@app.route('/transcribe', methods=['POST'])
def transcribe_audio():
if 'audio' not in request.files:
return jsonify({"error": "No audio file provided"}), 400
audio_file = request.files['audio']
# Save the uploaded file to a temporary location
with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as tmp_audio:
audio_file.save(tmp_audio.name)
tmp_path = tmp_audio.name
try:
# Run whisper.cpp to transcribe the audio
cmd = [
WHISPER_CPP_PATH,
"-m", MODEL_PATH,
"-f", tmp_path,
"--output-txt",
"--output-file", os.path.join(tempfile.gettempdir(), "output")
]
result = subprocess.run(cmd, capture_output=True, text=True)
if result.returncode != 0:
return jsonify({
"error": "Transcription failed",
"stderr": result.stderr
}), 500
# Read the output file
output_file = os.path.join(tempfile.gettempdir(), "output.txt")
with open(output_file, 'r') as f:
transcription = f.read()
return jsonify({
"transcription": transcription.strip()
})
finally:
# Clean up temporary files
if os.path.exists(tmp_path):
os.unlink(tmp_path)
output_file = os.path.join(tempfile.gettempdir(), "output.txt")
if os.path.exists(output_file):
os.unlink(output_file)
if __name__ == '__main__':
app.run(host='0.0.0.0', port=4002)