RealtimeSTT Releases

This release brings major infrastructure improvements including thread-safe IPC with SafePipe, audio normalization to -0.95 dBFS, and a complete callback overhaul to run asynchronously via helper threads, along with wake word and VAD enhancements.

Apr 19, 2025

This release enhances real-time responsiveness by intelligently pausing transcription when VAD detects silence, improves client connection robustness with WebSocket-based checks, and adds remote wake word delay configuration.

Apr 11, 2025

A minor update that expands the SQL generation capabilities by adding support for more query types.

Apr 10, 2025

Updates Docker and Docker Compose clients, disables the Console feature by default for security, and fixes YAML comment preservation.

Mar 30, 2025

This release introduces new VAD callbacks on_vad_start and on_vad_stop that trigger on voice activity presence, while reverting the functionality of on_vad_detect_start and on_vad_detect_stop back to triggering when the system starts/stops detecting for VAD presence.

Mar 23, 2025

This release introduces enhanced logging with a dedicated named logger, adds the ability to disable the Faster-Whisper VAD filter, improves audio worker debug logging, and fixes VAD callback timing issues.

Mar 21, 2025

Minimal release notes: "# RealtimeSTT 0.3.98 - minor fix for pypi wheel"

Mar 10, 2025

Mar 10, 2025

Improves the warmup process by utilizing an audio file.

Feb 15, 2025

Adds timing control parameters to adjust stop and resume timestamps for recordings.

Jan 23, 2025

Fixes stt-server compatibility and adds initial prompt support for the realtime model.

Dec 18, 2024

This release addresses a critical dependency issue that prevented the library from importing necessary components from the `faster_whisper` backend.

Dec 13, 2024

Minimal release notes: "- upgrade to 0.3.91 since 0.3.9 had issues on PyPi"

Dec 12, 2024

This update introduces batched transcription capabilities to improve processing efficiency, adding configurable batch size parameters for both standard and real-time modes.

Dec 11, 2024

The CLI interface has been significantly enhanced with new command-line flags for speech end detection, language selection, device listing, and device index configuration, alongside better piping support.

Nov 25, 2024

This release fixes a client termination issue and refactors the CLI commands to be shorter and more user-friendly.

Nov 3, 2024

Enhanced logging capabilities for both client and server, allowing detailed debugging and audio chunk inspection, along with a new parameter to fine-tune the initial listening delay.

Nov 2, 2024

A maintenance release focused on improving Linux compatibility and fixing bugs within the CLI and server components.

Oct 29, 2024

A minor maintenance update addressing server stability and bug fixes.

Oct 27, 2024

The server and client now support wake words, and microphone reconnection logic has been updated.

Oct 27, 2024