Spaces:
Running
Running
| import firebase_admin | |
| from firebase_admin import credentials | |
| from firebase_admin import firestore | |
| import io | |
| from fastapi import FastAPI, File, UploadFile | |
| from werkzeug.utils import secure_filename | |
| import speech_recognition as sr | |
| import subprocess | |
| import os | |
| import requests | |
| import random | |
| import pandas as pd | |
| from pydub import AudioSegment | |
| from datetime import datetime | |
| from datetime import date | |
| import numpy as np | |
| from sklearn.ensemble import RandomForestRegressor | |
| import shutil | |
| import json | |
| from transformers import AutoModelForQuestionAnswering, AutoTokenizer, pipeline | |
| from pydantic import BaseModel | |
| from typing import Annotated | |
| from transformers import BertTokenizerFast, EncoderDecoderModel | |
| import torch | |
| import random | |
| import string | |
| import time | |
| from fastapi import Form | |
| class Query(BaseModel): | |
| text: str | |
| code:str | |
| # device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') | |
| # tokenizer = BertTokenizerFast.from_pretrained('mrm8488/bert-small2bert-small-finetuned-cnn_daily_mail-summarization') | |
| # model = EncoderDecoderModel.from_pretrained('mrm8488/bert-small2bert-small-finetuned-cnn_daily_mail-summarization').to(device) | |
| from fairseq.checkpoint_utils import load_model_ensemble_and_task_from_hf_hub | |
| from fairseq.models.text_to_speech.hub_interface import TTSHubInterface | |
| import IPython.display as ipd | |
| models, cfg, task = load_model_ensemble_and_task_from_hf_hub( | |
| "facebook/fastspeech2-en-ljspeech", | |
| # 'facebook/fastspeech2-en-200_speaker-cv4', | |
| arg_overrides={"vocoder": "hifigan", "fp16": False} | |
| ) | |
| model = models[0] | |
| TTSHubInterface.update_cfg_with_data_cfg(cfg, task.data_cfg) | |
| generator = task.build_generator([model], cfg) | |
| from fastapi import FastAPI, Request, Depends, UploadFile, File | |
| from fastapi.exceptions import HTTPException | |
| from fastapi.middleware.cors import CORSMiddleware | |
| from fastapi.responses import JSONResponse | |
| app = FastAPI() | |
| app.add_middleware( | |
| CORSMiddleware, | |
| allow_origins=['*'], | |
| allow_credentials=True, | |
| allow_methods=['*'], | |
| allow_headers=['*'], | |
| ) | |
| # cred = credentials.Certificate('key.json') | |
| # app1 = firebase_admin.initialize_app(cred) | |
| # db = firestore.client() | |
| # data_frame = pd.read_csv('data.csv') | |
| async def startup_event(): | |
| print("on startup") | |
| audio_space="https://audiospace-1-u9912847.deta.app/upload" | |
| # @app.post("/code") | |
| # async def get_code(request: Request): | |
| # data = await request.form() | |
| # code = data.get("code") | |
| # global audio_space | |
| # print("code ="+code) | |
| # audio_space= audio_space+code | |
| import threading | |
| async def get_answer(q: Query ): | |
| text = q.text | |
| code= q.code | |
| N = 20 | |
| res = ''.join(random.choices(string.ascii_uppercase + | |
| string.digits, k=N)) | |
| res= res+ str(time.time()) | |
| filename= res | |
| t = threading.Thread(target=do_ML, args=(filename,text,code)) | |
| t.start() | |
| return JSONResponse({"id": filename}) | |
| return "hello" | |
| import requests | |
| import io | |
| import torch | |
| from scipy.io import wavfile | |
| import soundfile as sf | |
| import wave | |
| import audioop | |
| import io | |
| def do_ML(filename:str,text:str,code:str): | |
| sample = TTSHubInterface.get_model_input(task, text) | |
| wav, rate = TTSHubInterface.get_prediction(task, model, generator, sample) | |
| def compress_audio_wav_bytes(wav_bytes, output_format): | |
| # Load the WAV data from bytes | |
| audio_data, sample_rate = sf.read((wav_bytes)) | |
| # Compress and save the audio data with the specified output format | |
| output_bytes = io.BytesIO() | |
| sf.write(output_bytes, audio_data, sample_rate, format=output_format) | |
| # Retrieve the compressed audio data as bytes | |
| compressed_bytes = output_bytes.getvalue() | |
| return compressed_bytes | |
| wav_bytes = io.BytesIO() | |
| # Write the audio data to the byte stream | |
| sf.write(wav_bytes, wav.numpy(), rate, format='WAV', subtype='PCM_16') | |
| # Set the position of the byte stream to the beginning | |
| wav_bytes.seek(0) | |
| format = 'flac' # Specify the output format ('flac', 'mp3', etc.) | |
| wav_bytes = compress_audio_wav_bytes(wav_bytes, format) | |
| files = {'file': wav_bytes} | |
| # global audio_space | |
| url = code | |
| data = {"filename": filename} | |
| response = requests.post(url, files=files,data= data) | |
| print(response.text) | |
| if response.status_code == 200: | |
| print("File uploaded successfully.") | |
| # Handle the response as needed | |
| else: | |
| print("File upload failed.") | |