Hecho para agentes de voz que no pueden esperar 600ms

Soniox: Soniox STT Realtime v6

soniox/stt-rt-v6
activo stt 1 proveedor
Playground

Soniox's sixth-generation real-time speech-to-text model, wire-compatible with v5 and adding agent text input and agent voice echo cancellation for voice agents.

AutorSoniox
Idiomasmultilingual
Estadoactivo

Proveedores

Distintos proveedores pueden alojar el mismo modelo. Elige un proveedor cuando necesites un backend fijo, o deja que el enrutamiento elija entre ellos.

ProveedorSlugPor horaLatenciaiEn TTS, la latencia se mide como el tiempo hasta el primer byte de audio. En STT, se mide como el tiempo hasta el primer token.ModosDisponibilidad
Soniox 8 parámetros
soniox
$0,12
—
Streaming

Idiomas admitidos

Todavía no hay una lista de idiomas revisada para este modelo.

Funciones

Speech to text

Transcribes spoken audio to text.

speech_to_text
Streaming

Real-time streaming transcription over WebSocket.

streaming
Multilingual transcription

Supports 60+ languages.

multilingual_transcription
Speaker diarization

Identifies and separates multiple speakers.

speaker_diarization
Word-level timestamps

Returns word-level timestamps.

word_timestamps
Language identification

Identifies the spoken language per token upstream; AllModels does not expose it on this route yet.

language_identification
Speech translation

Translates speech in the same request at no extra charge upstream; AllModels does not expose it on this route yet.

translation
Semantic endpointing

Provides fast semantic end-of-turn detection.

endpointing
Agent text input

Accepts the voice agent's own replies on the same WebSocket via enable_agent_text_input upstream; AllModels does not expose it on this route yet.

agent_text_input
Agent voice echo cancellation

Filters the agent's own voice out of the transcript when agent text input is enabled upstream; AllModels does not expose it on this route yet.

agent_voice_echo_cancellation

Código

Cliente
Modo
Lenguaje

STT en streaming con el ElevenLabs SDK

Transmite audio del micrófono o de un archivo a través de la ruta de AllModels compatible con ElevenLabs.

import { ElevenLabsClient } from "elevenlabs";

const client = new ElevenLabsClient({
  apiKey: process.env.ALLMODELS_API_KEY,
  baseUrl: "https://api.allmodels.io/el"
});

const connection = await client.speechToText.realtime.connect({
  modelId: "soniox/stt-rt-v6",
  languageCode: "en"
}, {
  queryParams: {
    provider: {
      "only": [
        "soniox"
      ]
    }
  }
});

// Send audio frames to connection, then handle transcript events.