transcript

This commit is contained in:
suraj.shenoy.b@gmail.com
2025-01-25 18:40:30 -06:00
parent 9f8fbccdfe
commit fe8e1faed8
6 changed files with 177 additions and 93 deletions

View File

@@ -1,49 +0,0 @@
from fastapi import FastAPI, File, UploadFile, HTTPException
from fastapi.middleware.cors import CORSMiddleware
import whisper
import os
import tempfile
import logging
# Set up logging
logging.basicConfig(level=logging.INFO)
app = FastAPI()
model = whisper.load_model("turbo") # Load the model once for efficiency
app.add_middleware(
CORSMiddleware,
allow_origins=["*"], # Frontend origin (adjust as needed)
allow_credentials=True,
allow_methods=["*"], # Allow all HTTP methods (GET, POST, etc.)
allow_headers=["*"], # Allow all headers (Authorization, Content-Type, etc.)
)
@app.post("/transcribe")
async def transcribe_audio(file: UploadFile = File(...)):
# Check the file extension
file_extension = file.filename.split('.')[-1].lower()
if file_extension not in ["mp3", "wav", "flac", "m4a"]:
raise HTTPException(status_code=400, detail="Invalid audio file format. Only mp3, wav, flac, or m4a are supported.")
try:
# Save the uploaded file to a temporary location
with tempfile.NamedTemporaryFile(delete=False, suffix=f".{file_extension}") as temp_file:
temp_file.write(await file.read())
temp_path = temp_file.name
logging.info(f"Audio file saved at: {temp_path}")
# Transcribe the audio using Whisper
result = model.transcribe(temp_path)
transcription = result["text"]
# Clean up temporary file
os.remove(temp_path)
logging.info(f"Temporary file {temp_path} removed after transcription.")
return {"transcription": transcription}
except Exception as e:
logging.error(f"Error during transcription: {e}")
raise HTTPException(status_code=500, detail="Internal server error during transcription.")

View File

@@ -0,0 +1,128 @@
import type { NextApiRequest, NextApiResponse } from "next";
import { createReadStream, unlinkSync } from "fs";
import path from "path";
import { IncomingMessage } from "http";
import { config } from "dotenv";
import formidable, { File } from "formidable";
import { AxiosError } from "axios";
import { OpenAI } from 'openai';
// Load environment variables
config();
const OPENAI_API_KEY = process.env.OPENAI_API_KEY;
if (!OPENAI_API_KEY) {
throw new Error("OpenAI API key is missing. Set OPENAI_API_KEY in your .env.local file.");
}
// Initialize OpenAI client
const openaiClient = new OpenAI({
apiKey: OPENAI_API_KEY,
});
export const apiconfig = {
api: {
bodyParser: false, // Disable default body parsing
},
};
// Helper to parse multipart form data
async function parseMultipartForm(req: IncomingMessage): Promise<{ filePath: string; originalFilename: string }> {
const form = formidable({
multiples: false, // Single file upload
uploadDir: "/tmp", // Temporary directory
keepExtensions: true,
maxFileSize: 50 * 1024 * 1024, // 50 MB
});
return new Promise((resolve, reject) => {
form.parse(req, (err, fields, files) => {
if (err) {
reject(err);
return;
}
const file = files.file as File | undefined;
if (!file) {
reject(new Error("No file found in the upload."));
return;
}
resolve({
filePath: file.filepath,
originalFilename: file.originalFilename || "unknown",
});
});
});
}
// Main handler
export default async function handler(req: NextApiRequest, res: NextApiResponse) {
if (req.method !== "POST") {
return res.status(405).json({ error: "Method not allowed. Use POST." });
}
let filePath: string | null = null;
try {
// Parse file upload
const { filePath: tempFilePath, originalFilename } = await parseMultipartForm(req);
filePath = tempFilePath;
// Log file details
console.log("Uploaded file path:", filePath);
console.log("Original filename:", originalFilename);
// Validate file extension
const allowedExtensions = ["mp3", "wav", "m4a"];
const fileExtension = path.extname(originalFilename).toLowerCase().replace(".", "");
if (!allowedExtensions.includes(fileExtension)) {
unlinkSync(filePath);
return res.status(400).json({
error: `Invalid file format. Only ${allowedExtensions.join(", ")} are supported.`,
});
}
// Create file stream
const audioFile = createReadStream(filePath);
console.log("File stream created for:", audioFile.path);
// Send to OpenAI Whisper API
console.log("Sending file to OpenAI Whisper...");
const response = await (openaiClient as any).createCompletion({
model: "whisper-1",
file: audioFile,
});
console.log("OpenAI response:", response.data);
// Clean up temporary file
unlinkSync(filePath);
// Send response back to client
return res.status(200).json({ transcription: response.data.text });
} catch (error) {
console.error("Error during transcription:", error);
if (error instanceof AxiosError) {
console.error("OpenAI API error:", error.response?.data || error.message);
return res.status(error.response?.status || 500).json({
error: error.response?.data.error?.message || "OpenAI API Error.",
});
}
return res.status(500).json({
error: "An unexpected error occurred.",
});
} finally {
if (filePath) {
try {
unlinkSync(filePath);
} catch (err) {
console.error("Failed to clean up temporary file:", err);
}
}
}
}