SSI (Speech Super Intelligence)

usecase 1 - use in existing reposall models will be loaded from the existing repo and then the server will be started

pip install ssi

from fastapi import FastAPI
from stapesai_ssi.fastapi import StreamingWSRouter
from stapesai_ssi.types import StreamingDataChunk, NewClientConnected

app = FastAPI()

def asr_callback(data: StreamingDataChunk):
    print(data)

def new_client_callback(data: NewClientConnected):
    print(data)

streaming_ws_router = StreamingWSRouter(
    asr_callback=asr_callback,
    new_client_callback=new_client_callback,
    endpoint="/ws/transcribe"
)

app.include_router(streaming_ws_router)

usecase 2 - deploy as a docker server and then use in existing repos here SSI server is deployed as a docker container and then ssi_client can be used in existing repos

pip install ssi[client]

from fastapi import FastAPI
from stapesai_ssi.clients import StreamingClient
from stapesai_ssi.types import StreamingDataChunk

app = FastAPI()

def asr_callback(data: StreamingDataChunk):
    print(data)

def new_client_callback(data: NewClientConnected):
    print(data)

ssi_client = StreamingClient(
    server_host="localhost",
    server_port=8000,
    asr_callback=asr_callback,
    new_client_callback=new_client_callback,
)

app.include_router(ssi_client.fastapi_proxy_router)

usecase 3 - CLI (for local system as well as simple socket server) bc dimak nhi chal rha abhi

python setup.py sdist bdist_wheel

twine upload dist/* --verbose