gladiaio-sdk

What it is and what it does

Gladia SDK is a Python client library for the Gladia speech-to-text service. It provides both synchronous and asynchronous interfaces for transcribing audio—either from pre-recorded files (local paths, file objects, or HTTP URLs) or from live audio streams via WebSocket. The SDK handles authentication via API key, manages file uploads when needed, and polls for job completion on pre-recorded jobs. For live sessions, it establishes WebSocket connections and emits events (partial transcripts, final transcripts, errors, session lifecycle) that your code can listen to and act on.

The package depends on httpx for HTTP requests, websockets for real-time connections, pyee for event handling, and dataclasses-json and marshmallow for request/response serialization. It requires Python 3.10 or later. The API surface includes methods like `transcribe()` for pre-recorded audio, `prerecorded_async()` for async pre-recorded work, and `live()` / `live_async()` for real-time transcription, all accepting configuration for language, diarization, translation, and other audio intelligence features.

Use it for:

• Transcribe pre-recorded audio files (MP3, WAV, etc.) with language detection and optional translation
• Build real-time transcription into live applications using WebSocket-based streaming
• Extract speaker identity and turn-taking information via diarization in recorded or live audio
• Redact personally identifiable information from transcripts automatically during processing
• Integrate speech-to-text into async Python applications without blocking the event loop

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

A Python SDK for the Gladia speech-to-text API, supporting both pre-recorded audio transcription and live streaming transcription with features like diarization, translation, and PII redaction.

Yes, if you need to integrate Gladia's speech-to-text service into a Python application. The SDK is actively maintained, has low install friction, and provides both sync and async interfaces for flexibility. License treatment is unclear, so verify the terms before production use. No known security vulnerabilities as of the query date.

Install

gladiaio-sdk on PyPI

pip install gladiaio-sdk

uv add gladiaio-sdk

poetry add gladiaio-sdk

Installing gladiaio-sdk

Before you install

Low install friction with a pure Python wheel. Requires Python 3.10+. Active maintenance with a recent release 22 days ago. Five runtime dependencies (dataclasses-json, httpx, marshmallow, pyee, websockets) are all standard, well-maintained libraries.

Quickstart

from gladiaio_sdk import GladiaClient

gladia_client = GladiaClient(api_key="your-api-key")
transcription = gladia_client.prerecorded().transcribe(
    "audio.mp3",
    {"language_config": {"languages": ["en"]}},
)
print(transcription.result.transcription.full_transcript)

Python 3.10+ required. Gladia API key must be provided via api_key parameter or GLADIA_API_KEY environment variable.

Verify before relying

• Whether the Gladia API key and service are free or require paid subscription
• Supported audio formats beyond the wav/pcm example shown
• Rate limits or quotas on transcription requests
• Latency characteristics for live transcription

Package facts

Evidence: gladiaio_sdk-1.0.5-py3-none-any.whl

Tags

More Linguistic packages