Grok Voice Transcribe

Insanely accurate transcription for phone calls, meetings, videos, and podcasts.

Try It Free

View Docs

Try it live

Speak into your mic and watch the transcript appear in real time.

Batch for files, WebSocket for low-latency streaming.
Word-level timestamps, diarization, and multichannel support.

Click “Start Transcribing” to begin...

Fast, accurate transcription

Optimized for diverse audio environments and real-world conditions: telephony, background noise, and more.

Leads overall accuracy across enterprise domains.
Inverse text normalization for numbers, dates, and currencies.

Accuracy (higher is better)

Grok Voice Transcribe 2.0SpaceXAI

97.4%

Scribe v2ElevenLabs

96.7%

Gemini 3.5 TranscribeGoogle

96.5%

Nova 3Deepgram

93.2%

Simple, transparent pricing

Usage-based billing with no hidden fees.

A fraction of competitors' prices.
Pay only for what you use.

Cost per hour of audio

Grok Voice Transcribe 2.0SpaceXAI

$0.10

Scribe v2ElevenLabs

$0.22

Nova 3Deepgram

$0.26

Gemini 3.5 TranscribeGoogle

$0.30

Batch + streaming

REST API for transcribing file uploads, WebSocket for live transcription.

Drop audio to transcribe

meeting.mp3

How can I help you today

Inverse text normalization

Spoken numbers, dates, and currencies become written form.

Spoken

one hundred dollars due june twelfth at three thirty p.m.

Transcribed

$100 due June 12 at 3:30 PM

Keyterm boosting

Bias towards your product names and proper nouns.

"keyterm": [

"Grok",

"SpaceXAI",

"Understand The Unive

]

Word-level timestamps

Per-word start and end times for karaoke-style captions.

Nice0.92s

to1.14s

meet1.22s

you,1.44s

Jane.1.60s

Speaker diarization

Word-level speaker IDs to separate transcription for up to 20 speakers.

Speakers

Thanks for calling. How can I help?

I can't log into my account.

What's the email on file?

Multichannel

Transcribe up to 2 channels separately at the same time.

Agent

Customer

Multilingual

Enterprise-grade transcription for 25+ languages.

Language25+

EnglishSpanishFrenchGermanJapanesePortugueseItalianKoreanHindiArabicDutchTurkish

Filler word control

Strip uh, um, and er from text and the words array by default.

Spoken

I, uh, need the, um, invoice

Transcribed

I need the invoice

Ready to transcribe?

Power your batch or streaming transcription with frontier voice models.

Create API Key Try on Playground

import os
import requests

r = requests.post(
    "https://api.x.ai/v1/stt",
    headers={
        "Authorization": f"Bearer {os.environ['XAI_API_KEY']}",
    },
    files={"file": open("audio.mp3", "rb")},
    data={"model": "grok-voice-transcribe-2.0"},
)
print(r.json()["text"])

curl -X POST https://api.x.ai/v1/stt \
  -H "Authorization: Bearer $XAI_API_KEY" \
  -F model=grok-voice-transcribe-2.0 \
  -F file=@audio.mp3

import fs from "fs";

const form = new FormData();
form.append("model", "grok-voice-transcribe-2.0");
form.append(
  "file",
  new Blob([fs.readFileSync("audio.mp3")]),
  "audio.mp3",
);

const res = await fetch("https://api.x.ai/v1/stt", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.XAI_API_KEY}`,
  },
  body: form,
});
console.log((await res.json()).text);

Explore the full Grok Voice suite

Speech to Speech

Intelligent, real-time conversations with sub-second latency.

$0.08 / min

Speech to Text

Enterprise-grade transcription for meetings, videos, and more.

$0.10 / hr batch$0.20 / hr streaming

Text to Speech

Natural, expressive speech from text in 25+ languages.

$15.00 / 1M characters

Voice Cloning

Clone a voice from two minutes of speech and use it instantly.

30 Free Clones

Live Translation

Speak in one language and be heard in another.

Coming Soon