LMSupply.Transcriber
0.112.0
See the version list below for details.
dotnet add package LMSupply.Transcriber --version 0.112.0
NuGet\Install-Package LMSupply.Transcriber -Version 0.112.0
<PackageReference Include="LMSupply.Transcriber" Version="0.112.0" />
<PackageVersion Include="LMSupply.Transcriber" Version="0.112.0" />
<PackageReference Include="LMSupply.Transcriber" />
paket add LMSupply.Transcriber --version 0.112.0
#r "nuget: LMSupply.Transcriber, 0.112.0"
#:package LMSupply.Transcriber@0.112.0
#addin nuget:?package=LMSupply.Transcriber&version=0.112.0
#tool nuget:?package=LMSupply.Transcriber&version=0.112.0
LMSupply.Transcriber
Local speech-to-text transcription using Whisper models.
Features
- Zero-config: Models download automatically from HuggingFace
- GPU Acceleration: CUDA, DirectML (Windows), CoreML (macOS)
- MIT Licensed: OpenAI Whisper models
- 99+ Languages: Auto-detection and multilingual support
Quick Start
using LMSupply.Transcriber;
// Load the default model
await using var transcriber = await LocalTranscriber.LoadAsync("default");
// Transcribe audio
var result = await transcriber.TranscribeAsync("audio.wav");
Console.WriteLine(result.Text);
Console.WriteLine($"Language: {result.Language}");
Console.WriteLine($"Duration: {result.DurationSeconds}s");
Available Models
| Alias | Model | Size | WER | Description |
|---|---|---|---|---|
fast |
Whisper Tiny | ~150MB | 7.6% | Ultra-fast |
default |
Whisper Base | ~290MB | 5.0% | Balanced |
quality |
Whisper Small | ~970MB | 3.4% | Higher accuracy |
medium |
Whisper Medium | ~3GB | 2.9% | High quality |
large |
Whisper Large V3 | ~6GB | 2.5% | Highest accuracy |
english |
Whisper Base.en | ~290MB | 4.3% | English optimized |
parakeet-tdt |
Parakeet TDT 0.6B v3 (int8) | ~670MB | — | NVIDIA transducer, 25 European languages, no 30 s window, no translation. Opt-in only (never chosen by auto); Language is "und" unless you pass a hint — the model has no language-id output |
GPU Acceleration
Do not add ONNX Runtime packages (Microsoft.ML.OnnxRuntime*): LMSupply provisions the runtime itself, and a
second copy conflicts with it. ExecutionProvider.Auto uses CUDA when the CUDA 12 runtime and cuDNN 9 are
installed on the machine, CoreML on macOS, and the CPU otherwise. On Windows with an AMD or Intel GPU, ONNX
sessions run on the CPU (DirectML was removed in 0.67.0). See
GPU acceleration.
| Product | Versions Compatible and additional computed target framework versions. |
|---|---|
| .NET | net10.0 is compatible. net10.0-android was computed. net10.0-browser was computed. net10.0-ios was computed. net10.0-maccatalyst was computed. net10.0-macos was computed. net10.0-tvos was computed. net10.0-windows was computed. |
-
net10.0
- Concentus (>= 2.2.2)
- LMSupply.Core (>= 0.112.0)
- MathNet.Numerics (>= 5.0.0)
- NAudio (>= 3.0.1)
- NLayer.NAudioSupport (>= 3.0.0)
NuGet packages (2)
Showing the top 2 NuGet packages that depend on LMSupply.Transcriber:
| Package | Downloads |
|---|---|
|
FileFlux.Providers.LMSupply
LMSupply local ONNX model provider for FileFlux: document analysis (summarization, metadata extraction), embeddings, OCR, image captioning, and speech transcription — no API key required. |
|
|
PulsaSTT.SDK
Pulsa STT (Speech-to-Text) SDK for speech transcription |
GitHub repositories
This package is not used by any popular GitHub repositories.
| Version | Downloads | Last Updated |
|---|---|---|
| 0.115.1 | 0 | 10/8/2026 |
| 0.115.0 | 0 | 10/8/2026 |
| 0.114.0 | 48 | 10/8/2026 |
| 0.113.0 | 63 | 10/8/2026 |
| 0.112.1 | 50 | 10/7/2026 |
| 0.112.0 | 61 | 10/7/2026 |
| 0.111.0 | 76 | 10/7/2026 |
| 0.110.1 | 68 | 10/7/2026 |
| 0.110.0 | 63 | 10/7/2026 |
| 0.109.1 | 74 | 10/7/2026 |
| 0.109.0 | 107 | 10/7/2026 |
| 0.108.0 | 127 | 10/6/2026 |
| 0.107.0 | 91 | 10/6/2026 |
| 0.106.1 | 91 | 10/5/2026 |
| 0.106.0 | 107 | 10/5/2026 |
| 0.105.2 | 108 | 10/4/2026 |
| 0.105.1 | 103 | 10/4/2026 |
| 0.105.0 | 100 | 10/4/2026 |
| 0.104.0 | 105 | 10/3/2026 |
| 0.103.0 | 101 | 10/3/2026 |