38 lines
995 B
Python
38 lines
995 B
Python
"""FluentGerman.ai — Voice service (API provider + browser fallback)."""
|
|
|
|
import io
|
|
|
|
import openai
|
|
|
|
from app.config import get_settings
|
|
|
|
|
|
async def transcribe(audio_bytes: bytes, filename: str = "audio.webm") -> str:
|
|
"""Transcribe audio to text using OpenAI Whisper API."""
|
|
settings = get_settings()
|
|
client = openai.AsyncOpenAI(api_key=settings.llm_api_key)
|
|
|
|
audio_file = io.BytesIO(audio_bytes)
|
|
audio_file.name = filename
|
|
|
|
transcript = await client.audio.transcriptions.create(
|
|
model=settings.stt_model,
|
|
file=audio_file,
|
|
)
|
|
return transcript.text
|
|
|
|
|
|
async def synthesize(text: str) -> bytes:
|
|
"""Synthesize text to speech using OpenAI TTS API."""
|
|
settings = get_settings()
|
|
client = openai.AsyncOpenAI(api_key=settings.llm_api_key)
|
|
|
|
response = await client.audio.speech.create(
|
|
model=settings.tts_model,
|
|
voice=settings.tts_voice,
|
|
input=text,
|
|
response_format="mp3",
|
|
)
|
|
|
|
return response.content
|