Download
TranscriptionSuite is a free, open-source speech-to-text application designed for private, locally processed audio transcription. It supports recordings from microphones, system audio, and imported audio or video files. Users can choose among AI transcription engines, including Whisper, NVIDIA NeMo, SenseVoice, and VibeVoice. The software provides real-time dictation, long-form transcription, and automatic speaker identification. Its Audio Notebook organizes recordings through a searchable calendar interface with optional AI assistance. Users can export transcripts as text or subtitle files and use global shortcuts for dictation. TranscriptionSuite also supports GPU acceleration, remote connections, and compatible API integrations. Available for Windows, macOS, and Linux, it processes audio locally without mandatory cloud services.
Features
- Fully local speech transcription with offline processing and private audio storage
- Multiple AI transcription engines supporting Whisper, NeMo, SenseVoice, and VibeVoice
- Real-time dictation and long-form recording with live transcription previews
- Automatic speaker diarization for identifying different voices in recorded conversations
- Searchable Audio Notebook with calendar organization and optional AI-assisted analysis
- Audio and video importing with TXT, SRT, and ASS transcript exporting
Project Samples
