How long does audio transcription take?

Processing time depends on audio length and file size. A 1-minute audio clip typically transcribes in a few seconds, while a 10-minute recording may take 15 to 30 seconds. All processing completes directly in your browser session.

What is the maximum file size for transcription?

Konthora allows file uploads up to 100 MB in size and up to 10 minutes in recording duration per job.

Can I transcribe a video file, not just audio?

Yes. Konthora accepts video files in MP4, WebM, and MOV formats. FFmpeg extracts the audio track automatically before passing it to the Whisper model.

Is my audio file stored after transcription?

No. Uploaded media files and generated transcripts are temporary and automatically deleted from the server after 60 minutes.

Does transcription work for phone call recordings?

Yes, provided the audio is saved in a supported format (such as MP3, WAV, or M4A) and the voices are clear. Speech recorded with heavy background noise or phone compression may have reduced accuracy.

What export format should I choose?

Choose TXT for clean reading notes, SRT or VTT for video subtitles and closed captions, or JSON if you need raw segment timestamps for software integration.

How to Transcribe Audio to Text (Free, with Timestamps)

Quick Steps: Transcribe Audio in 4 Steps

1Go to Konthora’s free Audio-to-Text tool in your web browser.
2Upload your audio or video file (MP3, WAV, M4A, AAC, MP4, WebM, or MOV — up to 100 MB / 10 mins).
3Select your preferred timestamp grouping: sentence-level, paragraph-level, or word-level.
4Click Transcribe Audio and export your result as TXT, SRT, VTT, or JSON format.

Step-by-Step: How to Transcribe Audio on Konthora

Step 1 — Upload Your Audio or Video File

Open the Konthora Audio to Text workspace. Drag and drop your media file directly onto the upload area, or click to select a file from your computer.

Konthora supports MP3, WAV, M4A, AAC audio files and MP4, WebM, MOV video files. Files must be within 100 MB in size and 10 minutes in duration.

Step 2 — Choose Your Language and Timestamp Mode

Verify that your audio language is set to English. Next, choose how you want your timestamps formatted:

Sentence-level: Groups text into complete sentences with start and end timestamps. Ideal for reading.
Paragraph-level: Groups text by natural speech pauses. Ideal for meeting notes.
Word-level: Provides precise start and end timings for every individual word. Ideal for audio indexing.

Step 3 — Start Transcription

Click Transcribe Audio. Konthora uses the Whisper speech recognition model to convert your audio into accurate text with precise timestamps.

Step 4 — Download or Copy Your Transcript

Once transcription is complete, review your transcript in the browser. You can copy the text directly to your clipboard or download it in your format of choice: TXT, SRT, VTT, or JSON.

Transcribe Audio Free Now

Supported Audio and Video Formats

Konthora accepts 7 common media formats:

Audio Formats

MP3, WAV, M4A, AAC

Video Formats

MP4, WebM, MOV (audio track extracted)

For all formats, the maximum allowable file size is 100 MB and the maximum audio duration is 10 minutes.

Understanding Timestamp Options

For a detailed breakdown of timestamp modes and format compatibility, see our guide on transcription timestamps.

Sentence Timestamps

Sentence-level timestamps assign a start and end time to each full sentence. This produces clean, natural paragraphs suitable for reading articles or creating show notes.

Paragraph Timestamps

Paragraph timestamps group speech segments by natural pauses in conversation, ideal for long-form speeches and interview transcripts.

Word-Level Timestamps

Word-level timestamps record the exact start and end millisecond of every spoken word, useful for precise audio editing and video synchronization.

Exporting Your Transcript

TXT Format

Plain text format containing speech segments without timestamp headers.

SRT Format

SubRip Subtitle format with sequential numbers, timestamps, and lines.

VTT Format

WebVTT caption format supported natively by HTML5 video players.

JSON Format

Structured JSON containing segment objects with word-level timing data.

Example SRT Output

1
00:00:00,000 --> 00:00:03,500
Welcome to Konthora's audio transcription workspace.

2
00:00:03,800 --> 00:00:07,200
Upload your audio or video file to get accurate timestamps.

Tips for Better Transcription Results

Record with a dedicated microphone in a quiet room to reduce background noise.
Maintain a steady speaking distance from your microphone to keep volume levels consistent.
Avoid overlapping voices — single-speaker speech transcribes with highest accuracy.
Use high-bitrate MP3 or WAV files instead of heavily compressed voice memos.