Conçu pour les agents vocaux qui ne peuvent pas attendre 600ms
Inscription gratuiteConnexion

Google Gemini : Gemini 3.5 Transcribe

gemini/gemini-3-5-transcribe
actif stt 1 fournisseur
Playground

Google's dedicated speech-to-text model (launched 2026-08-26): pre-recorded audio transcription via the Interactions API with 85+ auto-detected languages, speaker diarization (up to 8 speakers), word-level timestamps, custom vocabulary, and verbatim or smart (disfluency-cleaned, formatted) transcripts.

AuteurGoogle Gemini
Langues83 langues
Statutactif

Fournisseurs

Plusieurs fournisseurs peuvent héberger le même modèle. Choisissez-en un lorsque vous avez besoin d’un backend fixe, ou laissez le routage choisir parmi eux.

FournisseurSlugPar heureLatenceiPour le TTS, la latence correspond au délai jusqu’au premier octet audio. Pour le STT, elle correspond au délai jusqu’au premier token.ModesDisponibilité
Google Gemini 5 paramètres
gemini
$0,3
—
Synchrone

Langues prises en charge

83 langues
Afrikaans af-ZA
Amharic am-ET
Arabic (Egypt) ar-EG
Armenian hy-AM
Assamese as-IN
Azerbaijani az-AZ
Belarusian be-BY
Bengali (Bangladesh) bn-BD
Bengali (India) bn-IN
Bosnian bs-BA
Bulgarian bg-BG
Bulgarian (Aromanian) rup-BG
Burmese my-MM
Catalan ca-ES
Cebuano ceb
Central Khmer km-KH
Croatian hr-HR
Czech cs-CZ
Danish da-DK
Dutch nl-NL
English (Great Britain) en-GB
English (India) en-IN
English (United States) en-US
Estonian et-EE
Farsi fa-IR
Filipino fil-PH
Finnish fi-FI
French fr-FR
Galician gl-ES
Georgian ka-GE
German de-DE
Greek el-GR
Gujarati gu-IN
Hausa ha-NG
Hebrew he-IL
Hindi hi-IN
Hungarian hu-HU
Icelandic is-IS
Indonesian id-ID
Italian it-IT
Japanese ja-JP
Javanese jv-ID
Kabuverdianu kea-CV
Kannada kn-IN
Kazakh kk-KZ
Korean ko-KR
Kyrgyz ky-KG
Latvian lv-LV
Lingala ln-CD
Lithuanian lt-LT
Macedonian mk-MK
Malay ms-MY
Malayalam ml-IN
Maltese mt-MT
Mandarin Chinese (Simplified) cmn-Hans-CN
Marathi mr-IN
Mongolian mn-MN
Nepali ne-NP
Norwegian nb-NO
Oriya or-IN
Polish pl-PL
Portuguese (Brazil) pt-BR
Portuguese (Portugal) pt-PT
Punjabi pa-IN
Punjabi (Gurmukhi script) pa-Guru-IN
Romanian ro-RO
Russian ru-RU
Serbian sr-RS
Sindhi (Arabic script) sd-Arab-IN
Slovak sk-SK
Slovenian sl-SI
Spanish (Latin America) es-419
Spanish (United States) es-US
Swahili (Kenya) sw-KE
Swedish sv-SE
Tajik tg-TJ
Telugu te-IN
Thai th-TH
Turkish tr-TR
Ukrainian uk-UA
Uzbek uz-UZ
Vietnamese vi-VN
Cantonese (Traditional) yue-Hant-HK

Fonctionnalités

Speech to text

Transcribes pre-recorded audio files to text via the Interactions API.

speech_to_text
Synchronous

Transcribes fully-uploaded audio in a single blocking Interactions API request (audio up to 1 hour; 30 minutes when diarization or word timestamps are enabled).

synchronous
Multilingual transcription

Automatically detects and transcribes 85+ languages, including code-switching; language_codes can pin BCP-47 locales.

multilingual_transcription
Word-level timestamps

timestamp_granularities ["word"] returns start/end offsets per word (caps audio at 30 minutes; incompatible with smart mode).

word_timestamps
Speaker diarization

diarization_mode "speaker" tags segments for up to 8 speakers (caps audio at 30 minutes; incompatible with smart mode).

diarization
Custom vocabulary

Biases recognition toward up to 1,000 supplied phrases (best results with up to 100).

custom_vocabulary
Smart transcription

Optional smart mode removes disfluencies, applies self-corrections and structured formatting/inverse text normalization (default is verbatim).

smart_transcription

Code

Client
Mode
Langage

STT en streaming avec l’ElevenLabs SDK

Diffuse l’audio du micro ou d’un fichier via la route AllModels compatible 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: "gemini/gemini-3-5-transcribe",
  languageCode: "en"
}, {
  queryParams: {
    provider: {
      "only": [
        "gemini"
      ]
    }
  }
});

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