Skip to content

Releases: Deenyoro/Scrivox

Release list

Scrivox v1.8.4

Scrivox v1.8.4 Latest
Latest

Choose a tag to compare

View all tags
@Deenyoro Deenyoro released this 29 Sep 23:41
v1.8.4
81cd053

Scrivox v1.8.4

GPU-accelerated transcription suite for Windows with NVIDIA CUDA. Transcribe audio/video, identify speakers with diarization, generate subtitles with speaker labels, analyze video keyframes, and summarize meetings.

Output formats: txt, md, srt, vtt, json, tsv
Subtitles: SRT and VTT with optional speaker labels (e.g. [Alice] Hello in SRT, <v Alice> voice tags in VTT)

Downloads

Variant Size What's included
Scrivox-Lite ~200-300 MB Transcription only. No diarization, vision, or summary.
Scrivox (Regular) ~500-800 MB All features (diarization, vision, summary). Downloads diarization models on first run — requires HuggingFace token.
Scrivox-Full - Not built for this release (the pipeline had no HF_TOKEN to download the diarization models).

Each variant has a SHA256SUMS-Scrivox*.txt with the SHA-256 checksums of its files. A variant without a -setup.exe had an installer of 2 GiB or more, over GitHub's release asset limit; use its .7z portable.

Installation

Installer (recommended): download the -setup.exe for your variant and run it. It installs Scrivox, creates shortcuts, registers an uninstaller, and makes Scrivox auto-detectable by SimpleReliableRecorder.

Portable:

  1. Download and extract the .7z archive
  2. Copy .env.example to .env and add your API keys
  3. Double-click Scrivox.exe for GUI, or use from command line

Requirements

  • Windows 10/11 (64-bit)
  • NVIDIA GPU with CUDA support

API Keys (add to .env or enter in GUI)

  • HF_TOKEN — Required for diarization in Regular variant (get one). Not needed for Full (models bundled) or Lite (no diarization).
  • OPENROUTER_API_KEY — Required for vision analysis and meeting summary (get one)

Built on GitLab CI.

Assets 8
Loading

Scrivox v1.8.1

Scrivox v1.8.1

Choose a tag to compare

View all tags
@github-actions github-actions released this 03 Aug 21:56
v1.8.1
b5fcc1e

Scrivox v1.8.1

GPU-accelerated transcription suite for Windows with NVIDIA CUDA. Transcribe audio/video, identify speakers with diarization, generate subtitles with speaker labels, analyze video keyframes, and summarize meetings.

Output formats: txt, md, srt, vtt, json, tsv
Subtitles: SRT and VTT with optional speaker labels (e.g. [Alice] Hello in SRT, <v Alice> voice tags in VTT)

Downloads

Variant Size What's included
Scrivox-Lite ~200-300 MB Transcription only. No diarization, vision, or summary.
Scrivox (Regular) ~500-800 MB All features (diarization, vision, summary). Downloads diarization models on first run — requires HuggingFace token.
Scrivox-Full ~1.5-2 GB All features with pre-bundled diarization models. No HuggingFace token needed for diarization.

Installation

Installer (recommended): download the -setup.exe for your variant and run it. It installs Scrivox, creates shortcuts, registers an uninstaller, and makes Scrivox auto-detectable by SimpleReliableRecorder.

Portable:

  1. Download and extract the .7z archive
  2. Copy .env.example to .env and add your API keys
  3. Double-click Scrivox.exe for GUI, or use from command line

Requirements

  • Windows 10/11 (64-bit)
  • NVIDIA GPU with CUDA support

API Keys (add to .env or enter in GUI)

  • HF_TOKEN — Required for diarization in Regular variant (get one). Not needed for Full (models bundled) or Lite (no diarization).
  • OPENROUTER_API_KEY — Required for vision analysis and meeting summary (get one)
Loading

Scrivox v1.8.0

Scrivox v1.8.0

Choose a tag to compare

View all tags
@github-actions github-actions released this 14 Jul 01:37
v1.8.0
ae70e05

Scrivox v1.8.0

GPU-accelerated transcription suite for Windows with NVIDIA CUDA. Transcribe audio/video, identify speakers with diarization, generate subtitles with speaker labels, analyze video keyframes, and summarize meetings.

Output formats: txt, md, srt, vtt, json, tsv
Subtitles: SRT and VTT with optional speaker labels (e.g. [Alice] Hello in SRT, <v Alice> voice tags in VTT)

Downloads

Variant Size What's included
Scrivox-Lite ~200-300 MB Transcription only. No diarization, vision, or summary.
Scrivox (Regular) ~500-800 MB All features (diarization, vision, summary). Downloads diarization models on first run — requires HuggingFace token.
Scrivox-Full ~1.5-2 GB All features with pre-bundled diarization models. No HuggingFace token needed for diarization.

Installation

  1. Download and extract the .7z archive
  2. Copy .env.example to .env and add your API keys
  3. Double-click Scrivox.exe for GUI, or use from command line

Requirements

  • Windows 10/11 (64-bit)
  • NVIDIA GPU with CUDA support

API Keys (add to .env or enter in GUI)

  • HF_TOKEN — Required for diarization in Regular variant (get one). Not needed for Full (models bundled) or Lite (no diarization).
  • OPENROUTER_API_KEY — Required for vision analysis and meeting summary (get one)
Loading

Scrivox v1.6.1

Scrivox v1.6.1

Choose a tag to compare

View all tags
@github-actions github-actions released this 22 Apr 18:09
v1.6.1
2314856

Scrivox v1.6.1

GPU-accelerated transcription suite for Windows with NVIDIA CUDA. Transcribe audio/video, identify speakers with diarization, generate subtitles with speaker labels, analyze video keyframes, and summarize meetings.

Output formats: txt, md, srt, vtt, json, tsv
Subtitles: SRT and VTT with optional speaker labels (e.g. [Alice] Hello in SRT, <v Alice> voice tags in VTT)

Downloads

Variant Size What's included
Scrivox-Lite ~200-300 MB Transcription only. No diarization, vision, or summary.
Scrivox (Regular) ~500-800 MB All features (diarization, vision, summary). Downloads diarization models on first run — requires HuggingFace token.
Scrivox-Full ~1.5-2 GB All features with pre-bundled diarization models. No HuggingFace token needed for diarization.

Installation

  1. Download and extract the .7z archive
  2. Copy .env.example to .env and add your API keys
  3. Double-click Scrivox.exe for GUI, or use from command line

Requirements

  • Windows 10/11 (64-bit)
  • NVIDIA GPU with CUDA support

API Keys (add to .env or enter in GUI)

  • HF_TOKEN — Required for diarization in Regular variant (get one). Not needed for Full (models bundled) or Lite (no diarization).
  • OPENROUTER_API_KEY — Required for vision analysis and meeting summary (get one)
Loading

Scrivox v1.6.0

Scrivox v1.6.0

Choose a tag to compare

View all tags
@github-actions github-actions released this 28 Feb 21:57
v1.6.0
da62822

Scrivox v1.6.0

GPU-accelerated transcription suite for Windows with NVIDIA CUDA. Transcribe audio/video, identify speakers with diarization, generate subtitles with speaker labels, analyze video keyframes, and summarize meetings.

v1.6.0 is the first stable release to include a working bundled pyannote/speaker-diarization-community-1

Output formats: txt, md, srt, vtt, json, tsv
Subtitles: SRT and VTT with optional speaker labels (e.g. [Alice] Hello in SRT, <v Alice> voice tags in VTT)

Downloads

Variant Size What's included
Scrivox-Lite ~200-300 MB Transcription only. No diarization, vision, or summary.
Scrivox (Regular) ~500-800 MB All features (diarization, vision, summary). Downloads diarization models on first run — requires HuggingFace token.
Scrivox-Full ~1.5-2 GB All features with pre-bundled diarization models. No HuggingFace token needed for diarization.

Installation

  1. Download and extract the .7z archive
  2. Copy .env.example to .env and add your API keys
  3. Double-click Scrivox.exe for GUI, or use from command line

Requirements

  • Windows 10/11 (64-bit)
  • NVIDIA GPU with CUDA support

API Keys (add to .env or enter in GUI)

  • HF_TOKEN — Required for diarization in Regular variant (get one). Not needed for Full (models bundled) or Lite (no diarization).
  • OPENROUTER_API_KEY — Required for vision analysis and meeting summary (get one)
Loading

Scrivox v1.5.15

Scrivox v1.5.15

Choose a tag to compare

View all tags
@github-actions github-actions released this 28 Feb 07:22
v1.5.15
4dd136a

Scrivox v1.5.15

GPU-accelerated transcription suite for Windows with NVIDIA CUDA. Transcribe audio/video, identify speakers with diarization, generate subtitles with speaker labels, analyze video keyframes, and summarize meetings.

v1.5.15 is the latest stable release to include a working bundled pyannote/speaker-diarization-3.1

Older releases can be ignored, v1.5.15 is being kept for compatibility due to the included model.

Output formats: txt, md, srt, vtt, json, tsv
Subtitles: SRT and VTT with optional speaker labels (e.g. [Alice] Hello in SRT, <v Alice> voice tags in VTT)

Downloads

Variant Size What's included
Scrivox-Lite ~200-300 MB Transcription only. No diarization, vision, or summary.
Scrivox (Regular) ~500-800 MB All features (diarization, vision, summary). Downloads diarization models on first run — requires HuggingFace token.
Scrivox-Full ~1.5-2 GB All features with pre-bundled diarization models. No HuggingFace token needed for diarization.

Installation

  1. Download and extract the .7z archive
  2. Copy .env.example to .env and add your API keys
  3. Double-click Scrivox.exe for GUI, or use from command line

Requirements

  • Windows 10/11 (64-bit)
  • NVIDIA GPU with CUDA support

API Keys (add to .env or enter in GUI)

  • HF_TOKEN — Required for diarization in Regular variant (get one). Not needed for Full (models bundled) or Lite (no diarization).
  • OPENROUTER_API_KEY — Required for vision analysis and meeting summary (get one)
Loading

Scrivox v1.5.14

Scrivox v1.5.14 Pre-release
Pre-release

Choose a tag to compare

View all tags
@github-actions github-actions released this 28 Feb 06:09
v1.5.14
04c97d3

Scrivox v1.5.14

GPU-accelerated transcription suite for Windows with NVIDIA CUDA. Transcribe audio/video, identify speakers with diarization, generate subtitles with speaker labels, analyze video keyframes, and summarize meetings.

Output formats: txt, md, srt, vtt, json, tsv
Subtitles: SRT and VTT with optional speaker labels (e.g. [Alice] Hello in SRT, <v Alice> voice tags in VTT)

Downloads

Variant Size What's included
Scrivox-Lite ~200-300 MB Transcription only. No diarization, vision, or summary.
Scrivox (Regular) ~500-800 MB All features (diarization, vision, summary). Downloads diarization models on first run — requires HuggingFace token.
Scrivox-Full ~1.5-2 GB All features with pre-bundled diarization models. No HuggingFace token needed for diarization.

Installation

  1. Download and extract the .7z archive
  2. Copy .env.example to .env and add your API keys
  3. Double-click Scrivox.exe for GUI, or use from command line

Requirements

  • Windows 10/11 (64-bit)
  • NVIDIA GPU with CUDA support

API Keys (add to .env or enter in GUI)

  • HF_TOKEN — Required for diarization in Regular variant (get one). Not needed for Full (models bundled) or Lite (no diarization).
  • OPENROUTER_API_KEY — Required for vision analysis and meeting summary (get one)
Loading

Scrivox v1.5.13

Scrivox v1.5.13 Pre-release
Pre-release

Choose a tag to compare

View all tags
@github-actions github-actions released this 28 Feb 05:08
v1.5.13
899048c

Scrivox v1.5.13

GPU-accelerated transcription suite for Windows with NVIDIA CUDA. Transcribe audio/video, identify speakers with diarization, generate subtitles with speaker labels, analyze video keyframes, and summarize meetings.

Output formats: txt, md, srt, vtt, json, tsv
Subtitles: SRT and VTT with optional speaker labels (e.g. [Alice] Hello in SRT, <v Alice> voice tags in VTT)

Downloads

Variant Size What's included
Scrivox-Lite ~200-300 MB Transcription only. No diarization, vision, or summary.
Scrivox (Regular) ~500-800 MB All features (diarization, vision, summary). Downloads diarization models on first run — requires HuggingFace token.
Scrivox-Full ~1.5-2 GB All features with pre-bundled diarization models. No HuggingFace token needed for diarization.

Installation

  1. Download and extract the .7z archive
  2. Copy .env.example to .env and add your API keys
  3. Double-click Scrivox.exe for GUI, or use from command line

Requirements

  • Windows 10/11 (64-bit)
  • NVIDIA GPU with CUDA support

API Keys (add to .env or enter in GUI)

  • HF_TOKEN — Required for diarization in Regular variant (get one). Not needed for Full (models bundled) or Lite (no diarization).
  • OPENROUTER_API_KEY — Required for vision analysis and meeting summary (get one)
Loading

Scrivox v1.5.12

Scrivox v1.5.12 Pre-release
Pre-release

Choose a tag to compare

View all tags
@github-actions github-actions released this 28 Feb 03:50
v1.5.12
4d0ebae

Scrivox v1.5.12

GPU-accelerated transcription suite for Windows with NVIDIA CUDA. Transcribe audio/video, identify speakers with diarization, generate subtitles with speaker labels, analyze video keyframes, and summarize meetings.

Output formats: txt, md, srt, vtt, json, tsv
Subtitles: SRT and VTT with optional speaker labels (e.g. [Alice] Hello in SRT, <v Alice> voice tags in VTT)

Downloads

Variant Size What's included
Scrivox-Lite ~200-300 MB Transcription only. No diarization, vision, or summary.
Scrivox (Regular) ~500-800 MB All features (diarization, vision, summary). Downloads diarization models on first run — requires HuggingFace token.
Scrivox-Full ~1.5-2 GB All features with pre-bundled diarization models. No HuggingFace token needed for diarization.

Installation

  1. Download and extract the .7z archive
  2. Copy .env.example to .env and add your API keys
  3. Double-click Scrivox.exe for GUI, or use from command line

Requirements

  • Windows 10/11 (64-bit)
  • NVIDIA GPU with CUDA support

API Keys (add to .env or enter in GUI)

  • HF_TOKEN — Required for diarization in Regular variant (get one). Not needed for Full (models bundled) or Lite (no diarization).
  • OPENROUTER_API_KEY — Required for vision analysis and meeting summary (get one)
Loading

Scrivox v1.5.11

Scrivox v1.5.11 Pre-release
Pre-release

Choose a tag to compare

View all tags
@github-actions github-actions released this 27 Feb 22:20
v1.5.11
1ee86b5

Scrivox v1.5.11

GPU-accelerated transcription suite for Windows with NVIDIA CUDA. Transcribe audio/video, identify speakers with diarization, generate subtitles with speaker labels, analyze video keyframes, and summarize meetings.

Output formats: txt, md, srt, vtt, json, tsv
Subtitles: SRT and VTT with optional speaker labels (e.g. [Alice] Hello in SRT, <v Alice> voice tags in VTT)

Downloads

Variant Size What's included
Scrivox-Lite ~200-300 MB Transcription only. No diarization, vision, or summary.
Scrivox (Regular) ~500-800 MB All features (diarization, vision, summary). Downloads diarization models on first run — requires HuggingFace token.
Scrivox-Full ~1.5-2 GB All features with pre-bundled diarization models. No HuggingFace token needed for diarization.

Installation

  1. Download and extract the .7z archive
  2. Copy .env.example to .env and add your API keys
  3. Double-click Scrivox.exe for GUI, or use from command line

Requirements

  • Windows 10/11 (64-bit)
  • NVIDIA GPU with CUDA support

API Keys (add to .env or enter in GUI)

  • HF_TOKEN — Required for diarization in Regular variant (get one). Not needed for Full (models bundled) or Lite (no diarization).
  • OPENROUTER_API_KEY — Required for vision analysis and meeting summary (get one)
Loading