Skip to content

MIGRATION.md

Latest commit

 

History

History
129 lines (82 loc) · 3.76 KB

MIGRATION.md

File metadata and controls

129 lines (82 loc) · 3.76 KB

Real-Time SDK Migration Guide

This guide helps users migrate from the legacy Speechmatics Real-Time Client (speechmatics-python) to the new Speechmatics Real-Time SDK (speechmatics-rt). The new SDK provides a cleaner API, better error handling, and improved performance with minimal dependencies.

Significant Changes

  • Restructured Async Client API: New async context manager pattern with AsyncClient
  • Improved event handler interface: Simplified event registration with .on() method
  • Better error handling: More specific exceptions and clearer error messages
  • Lightweight package: Minimal dependencies for faster installation and reduced conflicts
  • Enhanced authentication: Dedicated JWTAuth class for JWT token management
  • Streamlined configuration: Separate ConnectionConfig for WebSocket settings
  • URL and API key configuration: Allows loading URL and API key from environment variables

Breaking Changes

  • Import paths: speechmatics.client → speechmatics.rt
  • Client class: WebsocketClient → AsyncClient
  • Method names: .run() → .transcribe() for transcription operations
  • Event registration: .add_event_callback() → .on()
  • Event types: ServerMessageType.AddTranscript → ServerMessageType.ADD_TRANSCRIPT
  • Authentication: API key passed directly to client or through environment variable instead of ConnectionConfig
  • JWT setup: Separate JWTAuth class instead of ConnectionConfig.generate_temp_token parameter
  • Configuration: ConnectionConfig now only handles WebSocket connection settings
  • CLI not available: CLI will be released as a separate package

Installation

pip install speechmatics-rt

Basic Usage

Before

from speechmatics.client import WebsocketClient
from speechmatics.models import TranscriptionConfig

client = WebsocketClient("API-KEY")

conf = TranscriptionConfig(language="en")

await client.run(audio_stream, conf)

After

from speechmatics.rt import AsyncClient, TranscriptionConfig

async with AsyncClient("API-KEY") as client:

    conf = TranscriptionConfig(language="en")

    await client.transcribe(audio_stream, transcription_config=conf)

Event Callbacks Usage

Before

from speechmatics.client import WebsocketClient
from speechmatics.models import ServerMessageType, TranscriptionConfig

client = WebsocketClient("API-KEY")

def on_transcript(msg: dict):
    print(msg)

client.add_event_callback(ServerMessageType.AddTranscript, on_transcript)

conf = TranscriptionConfig(language="en")

await client.run(audio_stream, conf)

After

from speechmatics.rt import AsyncClient, TranscriptionConfig, ServerMessageType

def on_transcript(msg: dict):
    print(msg)

async with AsyncClient("API-KEY") as client:
    client.on(ServerMessageType.ADD_TRANSCRIPT, on_transcript)

    conf = TranscriptionConfig(language="en")

    await client.transcribe(audio_stream, transcription_config=conf)

JWT Usage

Before

from speechmatics.client import WebsocketClient
from speechmatics.models import TranscriptionConfig, ConnectionConfig

conn_config = ConnectionConfig(auth_token="API-KEY", generate_temp_token=True)

client = WebsocketClient(conn_config)

conf = TranscriptionConfig(language="en")

await client.run(audio_stream, conf)

After

ConnectionConfig is now exclusively used for websocket connection configuration.

from speechmatics.rt import AsyncClient, ConnectionConfig, TranscriptionConfig, JWTAuth

conn_config = ConnectionConfig(ping_timeout=60)

auth = JWTAuth("API-KEY")

async with AsyncClient(auth=auth, conn_config=conn_config) as client:

    conf = TranscriptionConfig(language="en")

    await client.transcribe(audio_stream, transcription_config=conf)