Methodology
This page is the source of truth for every accuracy, speed, and language claim on ScreenApp.io. Numbers come from our internal test corpus, our own production measurements, the Groq engineering case study, OpenAI's Whisper benchmarks, and xAI's published Grok Speech-to-Text benchmarks. Last refreshed: July 2026.
Transcription does not bet on a single vendor. ScreenApp routes each job to the provider best suited to the audio: source platform, length, channel layout, language. As of May 2026, the primary providers are:
Fallback providers (used when a primary provider is rate-limited or unavailable): Cloudflare Workers AI, Fireworks AI, Mistral, Baseten. A transcription job never fails because one vendor had an outage.
LLM layer for summarization, chat, and AI analysis: Google Gemini end-to-end. ScreenApp is NOT powered by GPT-4, ChatGPT, or Claude. The LLM that turns a raw transcript into structured summaries, chapter markers, action items, Q&A answers, and the chat interface is Gemini from start to finish.
Why multi-provider instead of single-vendor: each provider has a sweet spot. Whisper has the most languages, Groq makes it fast. Gemini is best at short audio and downstream LLM analysis. xAI Grok STT has the lowest published error rate on phone calls. Routing each job to its best fit beats picking one provider for everything.
Customer audio is never used to train any of these models. Audio is processed and deleted per your account's retention settings. Full data handling on the Trust Center.
In 2025, ScreenApp moved from a self-hosted Whisper deployment on AWS to Groq's inference infrastructure. Groq published the case study; the numbers below are from their engineering team's measurements.
| Metric | Before Groq | After Groq | Change |
|---|---|---|---|
| 20-minute transcription job | ~20 minutes | ~15 seconds | 20x faster |
| Per-minute transcription cost | baseline | 1/15th | 15x cheaper |
| Free-to-paid conversion | baseline | +30% | uplift |
| Annual recurring revenue (year-over-year) | baseline | +405% | growth attributed to the speed and cost gains |
Source: ScreenApp + Groq case study (groq.com).
What this means in practice: a 60-minute meeting completes in roughly 3 minutes end-to-end (transcription, diarization, summary generation). A 2-hour video processes in about 6 minutes. These are end-to-end times that include summarization and chaptering, not just raw transcription.
When a YouTube video already publishes a caption track, ScreenApp reads that track instead of transcribing the audio. No speech recognition runs at all, which is why it is fast. The text is written by YouTube, so none of the word error rates on this page describe it. Videos without a caption track go through the AI transcription path above, which is slower and is what our benchmarks measure.
| Path | Imports (n) | Median | p90 |
|---|---|---|---|
| Download + AI transcription (before, Jul 8 to 9 2026) | 61 | 41.1 sec | 159 sec |
| Caption import (after, Jul 16 to 17 2026) | 1,774 | 14.7 sec | 29 sec |
Method: both windows measure the same endpoint (import start to transcript indexed) on the same population, on production traffic. We state the gain as about 3x rather than a precise figure: the before-window sample is small (n=61), which puts the true median speedup somewhere between 1.9x and 3.7x at 95% confidence. The mean (3.58x) is inflated by a long tail of large videos, so the median is the honest central figure.
Two limits worth stating plainly. First, caption import currently serves free-tier YouTube imports; paid accounts still take the download and transcription path, which measured a 60.5 second median before and 44.7 seconds after (that difference is sampling noise, not a change in the product). Second, 14.7 seconds reflects the current implementation, which fetches captions with yt-dlp. Last refreshed: July 2026.
Word error rate (WER) counts substitutions + deletions + insertions per 100 reference words. Lower is better. Baseline figures below come from the published benchmarks for each underlying model; the per-condition rows are from our own April 2026 retest on 18 hours of public-domain audio per language across three conditions: studio (single speaker, treated room), conference (multi-speaker, room mic), and field (handheld phone mic, ambient noise).
| Language | Locale | Studio WER | Conference WER | Field WER | iPhone mic WER † | Speakers tested |
|---|---|---|---|---|---|---|
| English (US) | en-US | 4.2% | 7.8% | 12.4% | 14.9% | 4 |
| Spanish (Latin Am.) | es-419 | 5.1% | 9.2% | 14.6% | 17.5% | 3 |
| Spanish (Spain) | es-ES | 5.4% | 9.8% | 15.1% | 18.1% | 3 |
| Portuguese (BR) | pt-BR | 5.8% | 10.1% | 15.8% | 19.0% | 3 |
| Portuguese (PT) | pt-PT | 6.4% | 11.2% | 17.0% | 20.4% | 2 |
| French | fr-FR | 5.9% | 10.4% | 16.2% | 19.4% | 3 |
| German | de-DE | 6.1% | 10.8% | 16.5% | 19.8% | 3 |
| Italian | it-IT | 6.3% | 11.0% | 17.1% | 20.5% | 3 |
| Japanese | ja-JP | 7.8% | 13.5% | 19.8% | 23.8% | 2 |
| Korean | ko-KR | 7.5% | 13.1% | 19.2% | 23.0% | 2 |
| Mandarin (Simplified) | zh-CN | 7.9% | 14.0% | 20.4% | 24.5% | 3 |
| Hindi | hi-IN | 9.2% | 15.8% | 23.1% | 27.7% | 3 |
| Arabic (MSA) | ar | 9.6% | 16.2% | 24.0% | 28.8% | 2 |
| Russian | ru-RU | 6.8% | 11.5% | 17.4% | 20.9% | 3 |
| Indonesian | id-ID | 7.1% | 12.4% | 18.5% | 22.2% | 2 |
† iPhone mic WER is a projection, not a measurement. Computed as Field WER × 1.2 to account for the noise-cancellation, beam-forming, and codec losses that iPhone built-in microphones introduce on top of a handheld phone mic. The July 2026 retest will replace this column with measured numbers from an iPhone 14, iPhone 15, and iPhone 16 across the same 18 hours per language.
Diarization (attaching a speaker ID to each word) runs on whichever path fits the audio. Routing is automatic; you do not pick.
The pipeline reads the audio's channel layout, source platform, and content type, then picks the better path. Word-level granularity in both cases (not paragraph-level), so a one-sentence interjection in a multi-speaker meeting gets attributed correctly.
Sources: xAI Grok STT and TTS API announcement, Google Gemini.
ScreenApp supports 99 languages for transcription via Whisper Large-v3. A subset of 25 of those also supports speaker diarization via xAI Grok STT (marked with †).
Afrikaans, Albanian, Amharic, Arabic †, Armenian, Assamese, Azerbaijani, Bashkir, Basque, Belarusian, Bengali, Bosnian, Breton, Bulgarian, Burmese, Cantonese, Catalan, Chinese (Mandarin) †, Croatian, Czech, Danish, Dutch †, English †, Estonian, Faroese, Finnish, French †, Galician, Georgian, German †, Greek, Gujarati, Haitian Creole, Hausa, Hawaiian, Hebrew, Hindi †, Hungarian, Icelandic, Indonesian †, Italian †, Japanese †, Javanese, Kannada, Kazakh, Khmer, Korean †, Lao, Latin, Latvian, Lingala, Lithuanian, Luxembourgish, Macedonian, Malagasy, Malay, Malayalam, Maltese, Maori, Marathi, Mongolian, Nepali, Norwegian, Nynorsk, Occitan, Pashto, Persian, Polish, Portuguese †, Punjabi, Romanian, Russian †, Sanskrit, Serbian, Shona, Sindhi, Sinhala, Slovak, Slovenian, Somali, Spanish †, Sundanese, Swahili, Swedish, Tagalog †, Tajik, Tamil, Tatar, Telugu, Thai, Tibetan, Turkish †, Turkmen, Ukrainian, Urdu, Uzbek, Vietnamese †, Welsh, Yiddish, Yoruba.
† indicates languages with word-level speaker diarization via xAI Grok STT. Other languages are transcribed (text-only) without per-speaker attribution.
Whisper's language list source: github.com/openai/whisper. Grok STT language coverage source: xAI Voice docs.
ScreenApp ships native apps on iOS, macOS, and Android alongside the web app. The numbers below are pulled from each platform's canonical store listing or download endpoint and verified on May 15, 2026. This section is refreshed every release cycle.
This is the privacy declaration ScreenApp submits to Apple, rendered exactly as it appears on the App Store. Apple's listing is authoritative; if the table below ever diverges from the live App Store page, the App Store page wins and we will update this table within 7 days. ScreenApp declares no tracking data.
| Group | Category | Data types |
|---|---|---|
| Data Used to Track You | None. ScreenApp does not declare any tracking data. | |
| Data Linked to You | User Content | Photos or Videos, Audio Data |
| Identifiers | User ID | |
| Diagnostics | Performance Data | |
| Data Not Linked to You | Identifiers | Device ID |
| Contact Info | Email Address, Name | |
| Diagnostics | Crash Data, Other Diagnostic Data | |
Canonical source: the App Privacy section on the ScreenApp App Store page. Full data handling policies on the Trust Center.
ScreenApp-latest.dmg and always serves the current production build, so there is no separate version string to copy here.io.screenapp.screenapp_mobile.Rating and review counts move daily on the App Store and Google Play. The numbers on this page are point-in-time snapshots, dated above. The live store listings are always the authoritative source; if the divergence ever exceeds 0.2 stars or 10 percent of reviews, please flag it via the Trust Center contact form and we will refresh sooner.
A small sample of named, named-role customer reviews drawn from the public reviews page. These are samples, not the full corpus. Aggregate ratings on each app remain whatever the App Store and Google Play report (4.0 stars across 94 ratings on iOS, 3.9 stars across 678 ratings on Android), not the average of the three reviews below.
The desktop and mobile apps are fantastic. Recording meetings while I'm mobile has never been easier, and the dictation feature is a huge time-saver. It helps me develop faster and eliminates the guesswork of recalling meeting details.
Kelvin, Software Engineer
Our overall experience with ScreenApp has been nothing but pleasant. Their support is terrific, and ScreenApp is a great recording system.
Aaron, Verified Capterra User
Our team was drowning in client feedback until we found ScreenApp. Now we record every presentation and client call, and the AI summaries are spot-on. My team actually looks forward to review sessions now because everything is searchable and actionable.
Millie, Director
The numbers below are real production counts, pulled at build time from the same MongoDB cluster the dashboard reads from. They are not marketing rollups, not rounded, and not estimated. Refresh cadence: every deploy. Last pulled: July 21, 2026.
486,717
recordings processed
transcribed and analysed in production
1,960,773
speakers diarized
unique speaker turns identified across the corpus
292,368
AI Q&A sessions
questions asked against transcribed media
126,977
voice dictations
captured via browser, iOS, and Android
63,512
meeting-bot sessions
across Google Meet, Microsoft Teams, and direct integrations
350,401
analysed video metadata sets
meeting type, speakers, companies extracted
Recent activity (indexed proxy via videometainfo.createdAt): 975 recordings analysed in the last 24 hours, 7,153 in the last 7 days, 28,966 in the last 30 days. Daily rate of roughly 966 analyses per day.
Why videometainfo and not recordings directly: recordings._id is a UUID, so we cannot do indexed time-range queries on it. Each videometainfo doc maps 1:1 to a recording via the unique recordingId index, so the time-windowed counts above are a faithful proxy. Methodology and the open query module: below.
7,816,645 accounts registered as of July 21, 2026. Number reflects unique verified-email accounts in our production database, pulled at build time from the same source the dashboard reads from. The figure on this page refreshes every deploy, not on a quarterly cadence.
We do not publish round-number marketing claims like "2 million users" without the verifiable underlying count, on this page or elsewhere. If you ever see an inflated or undated user-count claim on a ScreenApp page, that's a content quality issue and we'd like to know: contact us via the Trust Center.
Every numeric claim on this page and across screenapp.io that depends on production data follows the same pipeline. Numbers are not curated, edited, or rounded for marketing.
GET /v2/site-data) that runs a small set of indexed aggregations and returns scalar counts. The query module is open inside the same repo at scripts/site-data-queries.ts.data/stats.db) and read synchronously by every page at static-generation time. Within a single deploy the numbers do not drift; between deploys they refresh.If you ever spot a number on the site that disagrees with a figure on this page, please flag it via the Trust Center. A divergence is a bug.
Two ways to use ScreenApp without paying upfront:
Current pricing and feature breakdown on the pricing page. What we do NOT offer: a recurring monthly free tier with X minutes/month, a "no credit card" trial of the paid plan, or unlimited free recordings. If you see those claims anywhere on this site, that's a content quality issue, please flag it via the Trust Center and we'll fix the source page.
SOC 2 Type 2 audited annually. 22 internal policies covering access control, data classification, secure development, and incident response. Continuous control monitoring.
Full live security posture, downloadable SOC 2 Type 2 report, and pre-filled security questionnaire at our Trust Center (trust.inc/screenapp).
Numbers on ScreenApp pages should match this page. If you find a feature page that contradicts these figures, that's a content quality bug we want to fix. Report it via the Trust Center contact form and we'll update the source page within 7 days.
Paste a URL or upload an audio or video file. See the actual accuracy on your content, not benchmarks on someone else's.
No charge for 7 days, cancel anytime • Aucune carte de crédit requise