Spaces:
Sleeping
Sleeping
File size: 1,619 Bytes
4cb1298 |
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 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 |
import os
from uuid import uuid4
import edge_tts
from groq import Groq
from dotenv import load_dotenv
load_dotenv()
client = Groq()
# ==================================================
# 🎙️ TEXT TO SPEECH (FIXED VOICE)
# ==================================================
DEFAULT_VOICE = "en-US-MichelleNeural"
async def TTS(
text: str,
output_dir: str = "tts_outputs",
rate: str = "+0%",
pitch: str = "+0Hz"
) -> str:
if not text.strip():
raise ValueError("Empty text")
os.makedirs(output_dir, exist_ok=True)
filename = f"{uuid4().hex}.mp3"
output_path = os.path.join(output_dir, filename)
communicate = edge_tts.Communicate(
text=text,
voice=DEFAULT_VOICE,
rate=rate,
pitch=pitch
)
await communicate.save(output_path)
return output_path
# ==================================================
# 🎧 SPEECH TO TEXT
# ==================================================
async def STT(audio_file):
os.makedirs("uploads", exist_ok=True)
file_path = f"uploads/{uuid4().hex}.wav"
with open(file_path, "wb") as f:
f.write(await audio_file.read())
with open(file_path, "rb") as f:
transcription = client.audio.transcriptions.create(
file=f,
model="whisper-large-v3-turbo",
response_format="verbose_json",
temperature=0.0
)
return {
"text": transcription.text,
"segments": transcription.segments,
"language": transcription.language
}
|