SaluteSpeech TTS Integration (Sberbank) for Speech Synthesis Salute
Speech TTS from Sberbank is a Russian neural speech synthesis system with a focus on Russian. Infrastructure in the Russian Federation, the possibility of on-premise deployment for secure circuits, GOST compliance. ### Voices and characteristics - Nec - neutral male - Bys - warm male - May - female - Tur - emotional male - Ost - official male - Pon - friendly female ### REST API integration
import requests
import base64
def get_access_token(client_id: str, client_secret: str) -> str:
response = requests.post(
"https://ngw.devices.sberbank.ru:9443/api/v2/oauth",
headers={
"Authorization": f"Basic {base64.b64encode(f'{client_id}:{client_secret}'.encode()).decode()}",
"RqUID": "unique-uuid-here",
"Content-Type": "application/x-www-form-urlencoded"
},
data={"scope": "SALUTE_SPEECH_CORP"},
verify=False # Потребуется корневой сертификат Сбер
)
return response.json()["access_token"]
def synthesize(text: str, voice: str = "Nec", token: str = None) -> bytes:
response = requests.post(
"https://smartspeech.sber.ru/rest/v1/text:synthesize",
headers={
"Authorization": f"Bearer {token}",
"Content-Type": "application/text",
"model": f"v4/\"{voice}\":emotion"
},
params={
"format": "wav16",
"voice": f"{voice}_24000"
},
data=text.encode("utf-8")
)
return response.content
```### Integration Features: The token expires every 30 minutes—we'll implement automatic updates via a background worker. Sber's SSL certificates aren't included in standard browser storage—we'll configure trust in the root CA. Price: upon request, corporate plans available. Timeframe: 2–3 days (including SSL-specific solutions).