--- title: "onespeech" url: https://recursiveai.org/onespeech/ description: "Speech recognition for .NET 10 in C#: transcribe audio, time words, label speakers and run live sessions on open-weight models, with no native inference runtime. Open source, MIT or Apache-2.0. 1.0.0." --- # onespeech Speech recognition for .NET, in C#. OneSpeech transcribes audio, times words, labels speakers and supports live sessions in managed .NET 10. **Open source**, MIT or Apache-2.0, made by Recursive AI. Version 1.0.0 is on NuGet, for .NET 10. Open-weight models run on [onemodel](https://recursiveai.org/onemodel/); [oneaudio](https://recursiveai.org/oneaudio/) decodes and resamples audio. No native inference runtime is required. OneSpeech is unaffiliated with other products or companies named OneSpeech. ```sh dotnet add package RecursiveAI.OneSpeech ``` - [Getting started](https://recursiveai.org/onespeech/docs/getting-started/) - [Docs](https://recursiveai.org/onespeech/docs/) - [GitHub](https://github.com/recursive-ai-oss/onespeech-csharp) - [NuGet](https://www.nuget.org/packages/RecursiveAI.OneSpeech) ## A quick look One call returns a timed, diarized, language-tagged transcript. Failed word timing is null, never invented, and a speaker label is anonymous and local to the document. The output shown is from a run on 1.0.0 from NuGet, on a three-second LibriSpeech recording whose reference text is “stuff it into you his belly counselled him”. The default file model, Whisper large-v3-turbo, heard “stuffered”. Getting started has the whole program. [Getting started](https://recursiveai.org/onespeech/docs/getting-started/) *Program.cs · C#* ``` Transcript t = Speech.Transcribe( "1089-134686-0001.flac", language: "en"); Console.WriteLine(t.Text); // stuffered into you, his belly counseled him. foreach (Word w in t.Words.Take(3)) Console.WriteLine($"{w.Start} {w.Text}"); // 00:00:00.5460000 stuffered // 00:00:00.8660000 into // 00:00:01.1860000 you, foreach (Turn turn in t.Turns) Console.WriteLine(turn.Speaker.Id); // speaker-1 File.WriteAllText("t.json", t.ToJson()); ``` ### Transcription Whisper large-v3-turbo is the file default; Whisper tiny to large-v3, Qwen3-ASR 0.6B and 1.7B and Moonshine are selected by name. Long files are chunked at speech boundaries with constant PCM memory. Output as text, JSON, SRT or VTT. [Transcribing files](https://recursiveai.org/onespeech/docs/transcribing-files/) ### Word timing Each word records how it was timed: native timestamps first, then Qwen3-ForcedAligner, omniASR-CTC and DTW. “Failed word timing is **null**, never invented.” `Speech.Align` times text you already have, up to 60 seconds. [Word timing](https://recursiveai.org/onespeech/docs/word-timing/) ### Speaker labels The default diarizer, Silero VAD with CAM++ embeddings and clustering, labels who spoke when, merged into the transcript. It doesn’t detect overlapping speech; Nemotron 3 Diarization is an opt-in, overlap-aware upgrade. [Diarization](https://recursiveai.org/onespeech/docs/diarization/) ### Live sessions `Speech.Listen` “emits replacing partials, immutable finals, endpoints and speaker turns on one audio clock.” It reads mono 16 kHz PCM you supply; it doesn’t capture a microphone. Moonshine streaming tiny is the live default. [Live sessions](https://recursiveai.org/onespeech/docs/live-sessions/) ### Speaker recognition `Speech.Enroll`, `Identify` and `Verify`, held in process memory only. Speaker embeddings are biometric data. Obtain informed consent, limit retention and access, and check applicable law, including Illinois BIPA and GDPR Article 9, before enrollment, identification or verification. [Speaker recognition](https://recursiveai.org/onespeech/docs/speaker-recognition/) ### Offline and captions Model packages put CAM++ English and Moonshine weights in your build, so a live session needs no download. `.Media` reads MP4, MKV and other containers and writes SRT, VTT, ASS and TTML; `.Cli` is the `onespeech` tool. [Command line and model packages](https://recursiveai.org/onespeech/docs/cli-and-model-packages/) ## What it is compared with Each row below is from the project’s release revalidation, on full LibriSpeech test-clean and test-other or 20,000 fixed speaker trials. “‘Within noise’ means the difference interval includes zero.” | OneSpeech, against | Verdict | |---|---| | Whisper large-v3-turbo and small, fp32, against openai-whisper 20250625 | Within noise | | Whisper large-v3-turbo and small, mixed-int8, against faster-whisper 1.2.1 (CT2 4.8.5 int8) | Within noise | | Moonshine streaming tiny, offline, against Transformers 5.22.2 | Within noise | | CAM++ English against ModelScope 1.40.2, and ECAPA-TDNN against SpeechBrain 1.1.2 (EER and minDCF) | Within noise | - **Word error rate.** Whisper large-v3-turbo fp32 scores 2.267% on test-clean against the reference’s 2.289%, and 4.275% on test-other against 4.331%. - **Speed.** “The tier-1 live default meets real time on one core on both machines, with 0 B/chunk of our working allocation. This does not imply that all file models run in real time on one core: turbo’s measured RTF is 1.712 on 60 seconds and 1.332 on 600 seconds.” - **Scope.** In the results’ own words, “these results are not a cross-hardware guarantee,” and “No speed comparison with a native implementation is made here.” [Measured results and limits](https://recursiveai.org/onespeech/docs/results/) ## Models by role, and their licences “Tier 1 is ungated MIT/Apache-2.0/BSD and may be redistributed and selected by default. Tier 2 loads only from pinned publisher files, never from our mirror or model packages, never by default.” | Role | Models | Tier / licence | Selection | |---|---|---|---| | File recognition | Whisper tiny/base/small/large-v3; large-v3-turbo | 1 / Apache-2.0; turbo MIT | Turbo default | | File recognition | Qwen3-ASR 0.6B / 1.7B | 1 / Apache-2.0 | By name; 1.7B accuracy option | | Live / edge recognition | Moonshine English tiny/base and streaming tiny/small/medium | 1 / MIT | Streaming-tiny live default | | Recognition | Parakeet-TDT-0.6B-v3 | 2 / CC-BY-4.0 | By name only | | Timing | Qwen3-ForcedAligner-0.6B; omniASR-CTC-300M | 1 / Apache-2.0 | Timing policy | | Voice activity | Silero VAD 6.2.3 | 1 / MIT | Embedded | | Diarization | Silero + CAM++ + spectral/AHC clustering | 1 / MIT + Apache-2.0 | Default; no overlap detection | | Diarization | Nemotron 3 Diarization | 2 / OpenMDW-1.1 | Explicit overlap-aware upgrade | | Segmentation | pyannote segmentation-3.0 / community-1 | 2 / MIT gated / CC-BY-4.0 gated | Local segmentation-3.0 + our overlap-aware clustering; community-1 reference only | | Speaker embeddings | CAM++ English/Chinese; ECAPA-TDNN | 1 / Apache-2.0 | CAM++ English default; Chinese for Chinese speech | [Tier-2 models and gates](https://recursiveai.org/onespeech/docs/tier-2-models/) ## Known limits at 1.0.0 - **Overlapping speech.** “The default clustering diarizer does not detect overlapping speech.” The Nemotron upgrade “does not separate simultaneous voices.” - **No microphone.** `Speech.Listen` consumes caller-supplied PCM. “It does not capture a microphone.” - **First use.** “The `model` argument selects a role, not the entire pipeline.” Transcription also needs a speaker embedder and, without native timings, aligners: “Even Whisper tiny can require several GB on first use.” `Speech.PrefetchAsync` prepares a machine ahead of time. - **Streaming and alignment.** `TranscribeStream` “does not perform whole-file speaker assignment,” and standalone `Align` “accepts at most 60 seconds.” [The README](https://recursiveai.org/onespeech/docs/overview/) ## Voices are biometric data “Speaker embeddings are biometric data. Obtain informed consent, limit retention and access, and check applicable law, including Illinois BIPA and GDPR Article 9, before enrollment, identification or verification.” The library does not persist embeddings; your application remains responsible for audio, transcripts, logs and any copies it keeps. In the README’s words, “This is not legal advice.” ## Status and packages - **Core**: `RecursiveAI.OneSpeech`: recognition, alignment, diarization, speakers, live sessions and the transcript JSON. It depends on [oneaudio](https://recursiveai.org/oneaudio/) and [onemodel](https://recursiveai.org/onemodel/) only. - **Media**: `RecursiveAI.OneSpeech.Media`: container input, SRT, VTT, ASS and TTML, and caption muxing, using [onemedia](https://recursiveai.org/onemedia/). - **Command line**: `RecursiveAI.OneSpeech.Cli`: the `onespeech` .NET tool. - **Model packages**: `RecursiveAI.OneSpeech.Models.CamPlusEnglish`, `.MoonshineTiny`, `.MoonshineBase` and `.MoonshineStreamingTiny`. Each covers only its own model, not every role `Transcribe` uses. - **Status**: 1.0.0, all seven packages on one version: the first release. - **.NET**: .NET 10, C#. NativeAOT and trim clean, no native code. MIT or Apache-2.0. ## Go further - [GitHub](https://github.com/recursive-ai-oss/onespeech-csharp) - [NuGet](https://www.nuget.org/packages/RecursiveAI.OneSpeech) - [Getting started](https://recursiveai.org/onespeech/docs/getting-started/) - [All docs](https://recursiveai.org/onespeech/docs/) - [First-use models](https://recursiveai.org/onespeech/docs/model-downloads/) - [Coming from Whisper.net](https://recursiveai.org/onespeech/docs/migrate-whisper-net/) - [Changelog](https://recursiveai.org/onespeech/docs/changelog/)