v1.0.2 Release - Streaming Audio Quality Improvements

🔊 Fixed: Streaming Audio Quality Issues

Eliminated clicks and micro-cuts in streaming TTS!

The most noticeable improvement in this release is the complete elimination of audio artifacts during streaming playback. We've achieved this through:

• Complete removal of audio trimming (TRIM_MS_FROM_END: 100ms → 0ms)
• Preservation of natural audio transitions between sentences
• Smart buffering algorithm with exponential growth for optimal performance

🚀 New Features & Improvements

Smart Buffering System

• Exponential buffer growth algorithm for adaptive streaming
• Multi-language sentence separators for better text segmentation
• Support for global punctuation patterns

Performance Optimizations

• Conditional audio processing to skip unnecessary operations
• Increased synthesis delay (100ms → 150ms) for better buffer synchronization
• Reduced CPU usage by eliminating redundant trimming operations

Enhanced Multi-language Support

• Comprehensive sentence separators: \n。.，,；;！!？?、
• Better handling of Asian languages and special punctuation
• Improved text segmentation for all supported languages

🎯 Who Should Update?

This update is highly recommended for all users experiencing:

• Audio clicks or pops during streaming TTS
• Unnatural pauses between sentences
• Buffering issues with long responses
• Quality issues with multi-language content

Version 1.0.1 - Bug Fixes on streaming

Fixed Issues

• 400 Bad Request Errors: Resolved API failures in streaming TTS by adding proper model parameter validation
• Voice Configuration: Fixed streaming TTS to use the same configured voice as regular TTS instead of always using the default
• Integration Setup: Fixed platform setup issues that were preventing proper integration loading
• Import Errors: Removed incorrect imports that were causing configuration failures

Technical Changes

• Enhanced API parameter validation to prevent empty model parameters
• Updated voice selection logic to be consistent between streaming and regular TTS
• Fixed integration setup and platform forwarding
• Added comprehensive error handling and debug logging
• Updated version to 1.0.1

v1.0.0

First public release of Deepgram TTS custom integration for Home Assistant.

• Supports Deepgram Aura-2 TTS
• Streaming TTS (Home Assistant 2025.7+)
• Voice and language selection
• HACS compatible