Voxa.Services.SpeechToSpeech 0.7.2-alpha

This is a prerelease version of Voxa.Services.SpeechToSpeech.

dotnet add package Voxa.Services.SpeechToSpeech --version 0.7.2-alpha

NuGet\Install-Package Voxa.Services.SpeechToSpeech -Version 0.7.2-alpha

This command is intended to be used within the Package Manager Console in Visual Studio, as it uses the NuGet module's version of Install-Package.

<PackageReference Include="Voxa.Services.SpeechToSpeech" Version="0.7.2-alpha" />

For projects that support PackageReference, copy this XML node into the project file to reference the package.

<PackageVersion Include="Voxa.Services.SpeechToSpeech" Version="0.7.2-alpha" />
                    

                            Directory.Packages.props

<PackageReference Include="Voxa.Services.SpeechToSpeech" />
                    

                            Project file

For projects that support Central Package Management (CPM), copy this XML node into the solution Directory.Packages.props file to version the package.

paket add Voxa.Services.SpeechToSpeech --version 0.7.2-alpha

The NuGet Team does not provide support for this client. Please contact its maintainers for support.

#r "nuget: Voxa.Services.SpeechToSpeech, 0.7.2-alpha"

#r directive can be used in F# Interactive and Polyglot Notebooks. Copy this into the interactive tool or source code of the script to reference the package.

#:package Voxa.Services.SpeechToSpeech@0.7.2-alpha

#:package directive can be used in C# file-based apps starting in .NET 10 preview 4. Copy this into a .cs file before any lines of code to reference the package.

#addin nuget:?package=Voxa.Services.SpeechToSpeech&version=0.7.2-alpha&prerelease
                    

                            Install as a Cake Addin

#tool nuget:?package=Voxa.Services.SpeechToSpeech&version=0.7.2-alpha&prerelease
                    

                            Install as a Cake Tool

The NuGet Team does not provide support for this client. Please contact its maintainers for support.

Voxa.Services.SpeechToSpeech

A full-duplex speech-to-speech composite for the Voxa pipeline (VRT-005) — the local-model peer of the cloud realtime composites (OpenAIRealtimeProcessor / AzureVoiceLiveProcessor).

SpeechToSpeechProcessor slots in exactly where those do — one processor that is the whole voice loop (model-owned VAD/turn-taking, full-duplex, native interruption) — but it is driven by an in-process (or sidecar) ISpeechToSpeechSession instead of a wire transport. It emits the same frame vocabulary the cloud composites do, so Studio's Talk view, the diagnostics hub, and any sink can't tell a third-party realtime API from a local model:

Direction	Frames
In	`AudioRawFrame` (user audio → the session), `InterruptionFrame` (barge-in → `session.CancelAsync`)
Out	`AudioRawFrame` (agent audio), `LlmTextChunkFrame`, `BotStartedSpeakingFrame`/`BotStoppedSpeakingFrame`, `UserStartedSpeakingFrame`/`UserStoppedSpeakingFrame`, `InterruptionFrame`, upstream `ErrorFrame`

The seam

ISpeechToSpeechSession (in Voxa.Speech.Abstractions) models speech-core's FullDuplexSpeechInterface: AppendUserAudioAsync, RespondAsync (a stream of SpeechToSpeechChunk carrying agent audio + text + the model's own speaking-edge events), SetVoiceAsync, SetSystemPromptAsync, ResetSessionAsync, CancelAsync. It is shaped deliberately parallel to what the cloud realtime processors do internally, so the composite is a true third member of that family rather than a parallel universe.

What ships vs. what's deferred

Ships: the seam + the composite processor, with frame parity tested against a fake session (no model).
Deferred (VRT-005 WS3 — spike-gated): a concrete ISpeechToSpeechSession. A real local speech-to-speech model (PersonaPlex / Moshi-class) is GB-scale and often GPU-only, so it lands behind a quality/feasibility spike, likely on a GPU or out-of-process host (reusing the VLS-002 / sidecar patterns). A cloud S2S provider could validate the seam first. Function-calling (ToolCallRequestFrame/ToolCallResultFrame) is a documented optional extension once a model that supports it lands.
Deferred (small follow-up): config-driven selection (a Voxa:Mode = SpeechToSpeech key, or a registered provider name, so UseDefaults() short-circuits to the composite). Today the composite is reached by constructing it directly / MapVoxaVoice(...).UseProcessor(...), exactly as a custom processor is.

Product	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.

Compatible target framework(s)

Included target framework(s) (in package)

Learn more about Target Frameworks and .NET Standard.

net10.0
- Microsoft.Extensions.Logging.Abstractions (>= 10.0.7)
- Voxa.Core (>= 0.7.2-alpha)
- Voxa.Speech.Abstractions (>= 0.7.2-alpha)

NuGet packages

This package is not used by any NuGet packages.

GitHub repositories

This package is not used by any popular GitHub repositories.