Upload a voice note, meeting, lecture or paste a YouTube link. Get clean, punctuated text in seconds. No login, no watermark, no limits.
MP3. Files up to 100 MB. Timestamps come back as a downloadable .srt./api/transcribe with a YouTube link or a base64 audio file. No key.English is the most accurate of all. High-resource languages are very strong, and every listed language is supported, especially with the Accurate model.
Send a POST request with a YouTube link or a base64-encoded audio file. Get plain text back. Free, no API key.
/api/transcribe
YouTube link
Fetch the transcript of any YouTube video. Send a JSON body with the youtube field.
| Param | Type | Description | |
|---|---|---|---|
| youtube | string | required | YouTube URL (youtube.com or youtu.be). |
| translate | string | optional | "1" to translate to English. |
curl -X POST https://ahm7xmakki.com/api/transcribe \ -H "Content-Type: application/json" \ -d '{"youtube":"https://youtu.be/VIDEO_ID"}'
const res = await fetch("/api/transcribe", { method: "POST", headers: { "Content-Type": "application/json" }, body: JSON.stringify({ youtube: "https://youtu.be/VIDEO_ID" }) }); const { text } = await res.json();
/api/transcribe
Audio file (base64)
Upload an audio file as a base64 data URL. Returns transcribed text and, if requested, an SRT subtitle file.
| Param | Type | Description | |
|---|---|---|---|
| audio | string | required | Base64 data URL of the audio file. |
| filename | string | optional | Original file name (helps the engine detect format). |
| mime | string | optional | Audio MIME type. |
| language | string | optional | Language code (e.g. en, ur) or auto. |
| model | string | optional | turbo, accurate or english. |
| translate | string | optional | "1" to translate to English. |
| timestamps | string | optional | "1" to also return an SRT string. |
{
"audio": "data:audio/mpeg;base64,SUQzBAAA…",
"filename": "clip.mp3",
"language": "auto",
"model": "turbo",
"timestamps": "1"
}
{
"text": "full transcript text …",
"language": "en",
"model": "turbo",
"srt": "1\n00:00:00,000 --> 00:00:04,120\n…"
}
Yes. Upload, transcribe and download at no cost. No account required, no watermark on the text.
Seconds for most files. Runs on a fast cloud engine, so you're not waiting minutes even for longer audio.
Up to 100 MB per file. Supported formats include MP3, WAV, M4A, AAC and FLAC. For fastest, cleanest results, upload an MP3.
99 in total. English is the most accurate. Major languages like Spanish, French, German, Arabic, Hindi and Chinese are very strong. Urdu and many others are supported and work best with the Accurate model.
Yes. Turn on Timestamps before transcribing and you also get a standard SRT subtitle file with timings.
Yes. Turn on Translate to English and the transcript comes back in English regardless of what language was spoken.
No. Your audio is sent to the transcription engine only to produce text, and is not saved on our servers.
Yes. POST /api/transcribe with either a YouTube URL or a base64-encoded audio file. No API key required.
New tools, updates, and exclusive resources delivered to WhatsApp.
TranScribe is a free audio-to-text service by AHM7xMakki — part of a 16-tool free API platform. Owned by AHM7. Built and managed by Makki. About us →
Your audio is sent through our server to a cloud speech engine only to produce the text. We don't save your audio or transcripts. Full policy →
For personal and lawful use. Only upload audio you have the right to transcribe. Fair-use limits apply. Full terms →
TranScribe by AHM7xMakki is a completely free audio-to-text transcription service. Upload MP3, WAV, M4A, AAC or FLAC audio, or paste a YouTube link, and get clean text back in seconds. Translate to English, export SRT subtitles, choose from Fast, Accurate or English-only models. No sign-up, no watermark, free REST API.