Skill UI

This SDK enables the development of real-time, conversational AI applications using Azure Cognitive Services. It supports bidirectional communication via WebSockets, allowing for continuous audio streaming (mic input) and receiving synthesized speech output. Ideal for building voice assistants, IVR systems, and interactive voice experiences.

This SDK provides a comprehensive interface for building real-time, bidirectional voice AI applications using Azure AI services. It supports streaming audio input/output via WebSocket, handling transcription, text-to-speech, and complex session management. Ideal for developing robust voice assistants and interactive AI agents in both Node.js and modern browser environments.

A comprehensive Java SDK for building robust, server-side call automation workflows using Azure Communication Services (ACS). It allows developers to implement sophisticated IVR systems, dynamic call routing, interactive media playback (Text-to-Speech, audio files), speech and DTMF recognition, call recording management, and participant handling. Ideal for building sophisticated customer service bots or automated notification systems.

A comprehensive Java SDK for building robust, server-side call automation workflows using Azure Communication Services (ACS). It allows developers to implement sophisticated IVR systems, dynamic call routing, interactive media playback (Text-to-Speech, audio files), speech and DTMF recognition, call recording management, and participant handling. Ideal for building sophisticated customer service bots or automated notification systems.

A Python implementation utilizing the Azure Speech-to-Text REST API for transcribing short audio files (up to 60 seconds). This method avoids the need for the dedicated Speech SDK, allowing developers to perform speech recognition simply by making HTTP POST requests. It supports both simple and detailed response formats and recommends chunked transfer for lower latency.

A comprehensive guide for diagnosing and fixing common Deepgram API errors. Covers HTTP error codes, WebSocket issues, optimal audio preparation using ffmpeg, and implementing robust retry patterns for reliable Speech-to-Text integration.

A comprehensive workflow for production-grade, pre-recorded Speech-to-Text (STT) transcription using the Deepgram API. This skill supports transcribing audio from URLs or local files, offering advanced features such as speaker diarization, automatic summaries, sentiment analysis, and topic detection. It also includes robust batch processing methods for handling large datasets.

This guide demonstrates how to implement robust, real-time, low-latency streaming transcription using Deepgram's WebSocket API. It covers advanced audio handling, including microphone input capture (via Sox), speaker diarization, utterance boundary detection, and implementing automatic reconnection logic. Essential for building professional voice AI applications, live captioning services, or voice interfaces.

This skill provides comprehensive methods to optimize Deepgram API usage and manage costs. It implements budget-aware transcription pipelines that preemptively check projected spending against set limits, utilizes audio preprocessing techniques (like silence removal via ffmpeg) to minimize billable minutes, and includes API calls to monitor current usage and spending against defined budgets. Ideal for cost-sensitive, production-grade deployments.

Implements best practices for secure audio transcription using Deepgram. This guide covers PII redaction (built-in and custom patterns), secure audio upload to S3 with encryption, managing data retention policies, and ensuring compliance with major regulations like GDPR and HIPAA. It provides a comprehensive pipeline for handling sensitive audio data in enterprise applications.

Collects comprehensive, sanitized debug evidence for Deepgram support and troubleshooting. This bundle gathers detailed system environment information (Node.js, Python), performs live API connectivity tests, analyzes audio file metadata using ffprobe, and logs request/response data. Essential for preparing detailed support tickets and reproducing bugs related to Deepgram integration.

This guide provides minimal working examples for Deepgram speech-to-text transcription. It demonstrates how to transcribe audio from a URL or a local file using both TypeScript and Python. Users can learn basic API patterns, implement advanced features like diarization (speaker identification), and compare various model options (e.g., nova-3, nova-2). Ideal for initial integration and testing.