ex_azure_speech

ExAzureSpeech is a community-driven Elixir implementation of the Microsoft Azure Cognitive Services Speech SDK. It enables Elixir applications to integrate speech recognition and synthesis functionalities using Azure's cloud infrastructure, making advanced speech AI accessible within the Elixir ecosystem.

Key Features

• Speech-to-Text — Converts audio streams to text with support for pronunciation assessment and detailed recognition results.
• Text-to-Speech — Synthesizes natural-sounding speech from text using Azure's neural voices and outputs to audio streams.
• Supervision Tree Integration — Recognizer and synthesizer modules are designed to be added to Elixir supervision trees for managed lifecycle.
• Pronunciation Assessment — Provides feedback on speech accuracy and fluency as part of the speech recognition process.
• Stream-Based Processing — Handles audio input and output using Elixir's streaming capabilities for efficient data handling.

Philosophy

The project aims to faithfully replicate all functionalities of the official Azure Speech SDK while adhering to Elixir's conventions and idioms, providing a native experience for Elixir developers.