Audio & Video Transcriber - Whisper ($0.06/min) avatar

Audio & Video Transcriber - Whisper ($0.06/min)

Pricing

from $42.00 / 1,000 audio minute transcribeds

Go to Apify Store
Audio & Video Transcriber - Whisper ($0.06/min)

Audio & Video Transcriber - Whisper ($0.06/min)

Audio transcription and speech to text with Whisper. Transcribe audio and video file URLs (MP3, WAV, M4A, MP4, MOV, WebM) to text, timestamped segments, SRT or VTT. Silence and music are filtered before decoding; every transcript carries quality flags. Pay per audio minute; failed URLs are free.

Pricing

from $42.00 / 1,000 audio minute transcribeds

Rating

5.0

(1)

Developer

Adam Rudd

Adam Rudd

Maintained by Community

Actor stats

2

Bookmarked

2

Total users

1

Monthly active users

10 hours ago

Last modified

Features

Limited permissions

Share

Audio Transcription: Speech to Text with Whisper (MP3, MP4, WAV to text, SRT, VTT)

Speech to text for your own audio and video files. Give direct file URLs (MP3, WAV, M4A, AAC, OGG, OPUS, FLAC, MP4, MOV, MKV, WebM) and get plain text, timestamped segments, SRT or WebVTT. It runs open-source Whisper (faster-whisper, MIT) on CPU inside the Actor, so your audio is not sent to any third-party speech service.

  • Input: direct links to media files (mp3, wav, m4a, aac, ogg, opus, flac, mp4, mov, mkv, webm and anything else FFmpeg can decode). Only the audio track of a video is used.
  • Output: one dataset item per URL: plain text plus, if you choose, timestamped segments, SRT or WebVTT. Every item says which engine and model (with its exact revision) produced it, whether silence filtering was on, and whether the transcript tripped a quality check.
  • Silence and music are filtered out before decoding (voice activity detection, on by default), because Whisper invents text such as "Thank you." on non-speech audio. See Measured.
  • Quality flags: each transcript is checked for the known failure shapes (repeated loops, one phrase repeated many times, uniform 30 s segments, long stretches with no text, a missing tail). A tripped check sets quality_degraded: true and says what and where in quality_flags.
  • Vocabulary prompt: give names and terms and they come out spelled your way (measured below).
  • Long files: audio is decoded in windows, so memory stays flat however long the file is (tested to 70 minutes).
  • Pricing: pay per audio minute. See Pricing.

How to transcribe an MP3 or video URL (no code)

  1. Paste one or more direct file links into Media URLs (the input form is pre-filled with a short public-domain LibriVox reading of Sonnet 130, so you can press Start straight away).
  2. Pick Output format: text, segments, srt or vtt.
  3. Optional: set Language if you know it, and put names or terms in Vocabulary prompt.
  4. Start. Each URL becomes one dataset item; export as JSON, CSV or Excel, or download the SRT/VTT text.

Use cases

  • Podcast, interview and lecture transcripts (text or timestamped segments).
  • Subtitles (SRT / WebVTT) for your own videos.
  • Transcripts with names spelled your way (vocabulary prompt; measured below).
  • Long recordings: memory stays flat (tested to 70 minutes).
  • Pipelines that must catch bad transcripts: quality_degraded and quality_flags on every item.

Premium model (opt-in)

Choose large-v3-turbo in the Whisper model field for our most accurate transcripts. On our shared test interview it made fewer errors than the leading Store transcriber (one 13-minute interview, measured 2026-10-23) and caught 66 of 67 technical terms; against our own standard model it also held up far better on noisy audio and music. Add a few names or terms in Vocabulary prompt (for example the video title) to help with spelling. It is charged on its own event, audio-minute-premium (see the Pricing tab); the default model stays on the standard price.

What it does NOT do

  • It does not download from web pages. The URL must return the media file itself; a URL that returns a web page fails with "URL returned a web page, not an audio or video file", and one that returns anything else that is not audio fails with "the URL did not return an audio or video file". Neither is charged.
  • No speaker diarisation and no word-level timestamps in this version.

Input

FieldDefaultNotes
mediaUrls(required)Direct media-file URLs
modelsmalltiny, base, small (standard price) or large-v3-turbo (premium price, see Pricing). small and large-v3-turbo are built into the image; tiny and base download in the run that uses them. (medium appears in the measurements below but is not offered.)
languageautoISO code such as en. Empty = detect. Auto-detect listens to only about the first 30 s of speech, so a file that opens with music or another language can be detected wrongly. Set it if you know it.
outputFormatsegmentstext, segments, srt or vtt. text is always included.
vocabulary_promptemptyNames and terms, spelled as you want them. Only the last 224 tokens are used (Whisper's limit): put the important terms at the end. A prompt that does not match the audio can bleed into the transcript.
vadFiltertrueSilero voice-activity detection (faster-whisper's vad_filter).
vadThreshold, vadNegThreshold, vadMinSpeechDurationMs, vadMaxSpeechDurationS, vadMinSilenceDurationMs, vadSpeechPadMs0.5, auto, 0, none, 2000, 400faster-whisper's own defaults, each explained in the input form
engineself_hostedThe only engine: Whisper inside the Actor.
windowSeconds600Decoding window, 30 to 3600
beamSize51 = greedy (faster)
carryContexttruePrompt each window with the previous window's last ~200 characters. Keeps punctuation and quieter speakers across window boundaries (measured on a 2-speaker interview: 3.1% vs 5.4% WER off)
maxDownloadMb / downloadTimeoutSecs / cpuThreads2000 / 300 / 0 (auto)

Output

{
"url": "https://.../episode-12.mp3",
"status": "ok",
"engine": "self_hosted",
"model": "small",
"modelRevision": "Systran/faster-whisper-small@536b0662742c02347bc0e980a01041f333bce120",
"vad": {"enabled": true, "threshold": 0.5, "min_speech_duration_ms": 0, "min_silence_duration_ms": 2000,
"speech_pad_ms": 400, "speechSeconds": 797.43},
"quality_degraded": false,
"quality_flags": [],
"language": "en",
"durationSeconds": 802.54,
"transcribedSeconds": 802.54,
"chargedUnits": 14,
"text": "One dollar and eighty-seven cents. That was all. ...",
"segments": [{"start": 0.0, "end": 3.6, "text": "One dollar and eighty-seven cents."}]
}

status: ok; partial (your maximum charge was reached or the run was aborted: the transcript covers what was paid for); failed (404, a web page or non-media file, or a decoding error part-way: then the transcript covers only the audio before the error and the error says so); rejected (not an http(s) URL, or a private address); not_processed (did not fit in your maximum charge, or the run had stopped).

quality_flags entries are {"gate", "detail"}. Gates: loops, repeats, uniform, gaps, coverage (thresholds from an established transcription checklist: 3 identical segments in a row, one text more than 25 times, 10+ segments of 28–31 s, more than 8 s with no text, the last segment ending 30 s or more before the audio does), plus decode_errors and window_failed. quality_degraded is null for items that were never transcribed. gaps also trips on real pauses longer than 8 s (for example 90 s of silence between two talks); the detail says where, so you can judge.

An item larger than Apify's dataset limit (9 MB of JSON, Apify docs) keeps its metadata in the dataset and stores the transcript in the run's key-value store, linked by transcriptKey / transcriptUrl.

How much does it cost to transcribe audio?

You pay per started minute of audio, rounded up per file. No start fee. Rejected and failed URLs are not charged. There are two prices: the standard models (tiny, base, small) are charged on the audio-minute event and the premium model (large-v3-turbo) on the audio-minute-premium event. The current prices for every Apify plan are on the Pricing tab. On the Free plan on 2026-11-09 they were $0.06 per minute (standard) and $0.0674 per minute (premium).

Example (Free plan, 2026-11-09 prices): a 60-minute podcast with small = 60 x $0.06 = $3.60.

If you set a maximum cost per run, a file that does not fit is skipped without charge (not_processed); if the limit is reached mid-file you get the transcript up to the last paid minute (partial). Larger models take longer.

Reliability (measured)

A small owner test run (a short clip, tiny model) runs every day (ops/apify_health.py): 25 of 25 passed between 2026-10-18 and 2026-11-09.

Measured (this Actor's own tests, 2026-09-30)

One shared 16-vCPU KVM host (Haswell, no AVX-512), often busy with other work. Every figure is one run (n=1), and Apify's hardware differs. WER uses Whisper's English normaliser.

Accuracy depends heavily on the audio

Word error rate by condition and model, using the Actor's default settings (600 s windows, beam 5, VAD on, language en, no vocabulary prompt). On this set, small (the default) ranged from 0.3% on clean synthetic speech to 20.7% on a clip full of unusual names. Real recordings (phone calls, crosstalk, heavy accents, poor microphones) were not tested and may be worse.

Conditiontinybasesmallmediumlarge-v3-turbo
TTS, clean (6 voices)1.4%0.7%0.3%0.3%0.6%
LibriVox story, one reader5.1%3.7%2.0%1.7%1.1%
LibriVox sonnet, 17 readers pooled [range]9.2% [4.0-16.9]6.7% [3.2-12.1]5.0% [3.2-8.1]*8.4% [3.2-32.3]*3.6% [1.6-7.3]*
TTS + pink noise, 20 dB1.2%1.2%0.4%0.3%0.2%
TTS + pink noise, 10 dB2.2%1.4%0.4%0.3%0.2%
TTS + pink noise, 5 dB4.4%3.5%0.9%0.9%0.6%
TTS + pink noise, 0 dB18.5%9.9%4.3%2.2%1.8%
TTS + music, 10 dB1.4%0.9%16.5%*0.2%0.2%
TTS + music, 0 dB9.7%*7.9%*7.2%10.7%*10.2%*
LibriVox story + pink noise, 10 dB7.9%5.4%3.0%1.7%1.5%
LibriVox story + music, 10 dB6.9%4.9%2.7%2.2%1.5%
TTS + competing talker, 10 dB11.1%4.6%1.1%0.6%4.5%
TTS through a phone channel10.9%*8.8%*7.8%7.0%*7.2%*
Unusual proper nouns, no prompt29.6%28.1%20.7%18.5%20.7%

WER in percent, lower is better; VAD on, n=1 run per cell. * = at least one quality gate tripped on that cell (see quality_flags in results/accuracy.jsonl).

How the set was built (only audio we have rights to):

  • LibriVox readings are public domain: The Gift of the Magi (one reader, 13.4 min) and Shakespeare's Sonnet 130 read by 17 different volunteer readers (different voices, accents and microphones, 56–91 s each). The sonnet reference is the Project Gutenberg text; its "damask'd" normalises to "damask would", so each reader carries up to two unavoidable errors (the sonnet figures are upper bounds). Readers' spoken introductions are not scored.
  • TTS: our own 5.5-minute meeting script read by six Kokoro-82M voices (two British, four American). Clean synthetic speech is the easy case.
  • Noise and music were synthesised (pink noise; a chord-and-melody loop) and mixed at the stated signal-to-noise ratio (speech level over its non-silent frames vs. the added sound). Talker: the LibriVox reading mixed 10 dB under the TTS speech. Phone: a synthetic phone channel (300–3400 Hz, 8 kHz, 8-bit mu-law).
  • tech_clean: a 1-minute Kokoro clip with 15 invented or rare proper nouns (24 occurrences), scored without a vocabulary prompt. With the prompt, see below.
  • A notable failure: small at music 10 dB dropped a whole ~120-word paragraph (and inserted a near-duplicate sentence). This happened with VAD on and with VAD off. The gaps check flagged it.

Silence and music (VAD)

Synthetic test file: 120 s of music, 45 s of speech, 90 s of digital silence, 45 s of speech. Segments whose midpoint falls in the music or silence are invented text:

modelVAD offVAD on (default)
large-v3-turbo22 invented segments ("Thank you." ×6 in 30 s blocks, then invented sentences such as "What do we do with the new site?"), loops flag0
small3 × "you" in the silence0
tiny9 ("Thanks for watching." ×2, then repeated real sentences placed in the silence)0

Pure silence (60 s): "you" ×2 with VAD off for every model, nothing with VAD on. Pure music (120 s): large-v3-turbo wrote "Thank you." ×4 with VAD off, nothing with VAD on. The cost: on the clean LibriVox reading, small scored 1.77% WER with VAD off and 1.96–2.0% with it on.

Vocabulary prompt

One-minute clip with 15 unusual names and terms (24 occurrences), e.g. "Oleksandra Wyrzykowski", "Zephyrine", "Hraunfoss". Exact spelling counts.

modelno promptwith prompt (80 tokens)terms first, then filler (464 tokens, cut to last 224)filler first, terms last (465 tokens, cut to last 224)prompt words in unrelated audio
tiny1/15 terms, 2/28 occurrences, WER 29.6%9/15 terms, 16/28 occurrences, WER 17.8%1/15 terms, 2/28 occurrences, WER 31.1%10/15 terms, 18/28 occurrences, WER 13.3%0
base0/15 terms, 0/28 occurrences, WER 28.1%13/15 terms, 25/28 occurrences, WER 7.4%0/15 terms, 0/28 occurrences, WER 32.6%12/15 terms, 23/28 occurrences, WER 8.9%0
small2/15 terms, 4/28 occurrences, WER 20.7%14/15 terms, 26/28 occurrences, WER 1.5%3/15 terms, 6/28 occurrences, WER 23.0%14/15 terms, 27/28 occurrences, WER 3.7%0
medium3/15 terms, 6/28 occurrences, WER 18.5%15/15 terms, 28/28 occurrences, WER 0.0%3/15 terms, 6/28 occurrences, WER 18.5%15/15 terms, 27/28 occurrences, WER 0.7%0
large-v3-turbo4/15 terms, 8/28 occurrences, WER 23.0%14/15 terms, 26/28 occurrences, WER 2.2%5/15 terms, 11/28 occurrences, WER 12.6%15/15 terms, 28/28 occurrences, WER 5.9%0

Terms first then filler: the prompt is over 224 tokens, so the terms are cut off and it behaves like no prompt. Filler first, terms last: the terms survive the cut. The 5th column counts prompt words that appeared in a different recording containing none of them.

The same prompt, applied to a different recording with none of those words, put 0 of them into the transcript for all five models (the bleed test). That is one test, not proof that bleed cannot happen.

Memory

Peak memory depends on the window length and the model, not on file length (70-minute file, small, 600 s windows: 1.6 GB for the whole Actor process). large-v3-turbo peaked at 2.1 GB on the Actor's own path, so it needs the 4 GB setting; it does not fit 2 GB.

FAQ

Can I call it from code or an AI agent? Yes, like any Apify Actor: the Apify API, the Python and JavaScript clients, the Apify MCP server, and the Make, Zapier and n8n integrations.

Does it take YouTube or web-page links? No. The URL must return the media file itself. A web page fails with a message and is not charged. For YouTube captions use YouTube Transcript.

Speaker labels or word timestamps? Not in this version.

Rights and responsible use

Only submit media that you own or are licensed to process, with any consent the recording requires.

About this Actor

Built, tested and maintained with an AI coding agent working for its publisher. Dependencies: faster-whisper 1.2.1 (MIT), CTranslate2, PyAV 15.1.0 (bundled FFmpeg), Silero VAD (via faster-whisper), OpenAI Whisper weights converted to CTranslate2 (MIT; revisions pinned and checked at build), Apify SDK for Python 4.0.2.

Feedback

Tried it? Tell us what worked and what didn't on the Issues tab: wrong words, a file that failed, a format you need. We read every issue and reply.