Spaces:
Sleeping
Sleeping
File size: 1,404 Bytes
ea25f01 9ab1a01 ea25f01 9ab1a01 ea25f01 9ab1a01 ea25f01 9ab1a01 ea25f01 9ab1a01 ea25f01 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 |
from fastapi import FastAPI, UploadFile, File, Response
from transformers import pipeline
import librosa
from deep_translator import GoogleTranslator
import io
app = FastAPI()
# print("Loading Speech Recognition")
# print("Speech Recognition Loaded")
print("Loading translator")
translator = GoogleTranslator(source='ku', target='fr')
print("Translator loaded")
# print("Loading tts")
# print("TTS loaded")
def speech2text(audio_data: bytes):
audio_array, _ = librosa.load(io.BytesIO(audio_data), sr=16000)
pipe = pipeline("automatic-speech-recognition", model="Akashpb13/xlsr_kurmanji_kurdish")
output = pipe(audio_array)
return output["text"]
def text2speech(text:str):
tts = pipeline("text-to-audio", model="roshna-omer/speecht5_tts_krd-kmr_CV17.0")
output = tts(text)
return output["audio"]
@app.post("/transcribe")
async def transcribe(file: UploadFile = File(...)):
audio_data = await file.read()
text_output = speech2text(audio_data)
translated = translator.translate(text_output)
return {"text": text_output, "translation": translated}
@app.post("/transcribe_audio")
async def transcribe_and_return_audio(file: UploadFile = File(...)):
audio_data = await file.read()
text_output = speech2text(audio_data)
audio_output = text2speech(text_output)
return Response(content=audio_output, media_type="audio/wav") |