Cleanvoice AI
AI audio enhancer, filler word cutter, and background noise remover for podcasters and developers (2 benchmarked capabilities: Voice isolation & denoising, Audio transcription)
Product node overview & trust baseline
Operator provenance, privacy baseline and site-wide pricing belong directly to the product node and are inherited across linked task edgesCleanvoice AI, operated by Sigmoid Creativity S.R.L. (Romania) under a "Subscription + credits" model, has 2 published task assessments on RealTest (Voice isolation & denoising 91/100, Audio transcription 70/100). In voice isolation & denoising, cleanvoice AI offers spoken-word denoising and enhancement with a 30-minute no-card trial; in practice, guest trials face strict quota limits, triggering 'You have reached the limit for guest uploads' on the 15s test anchor, withholding publication. In audio transcription, provides online processing capabilities.
Romania · Domain registered in 2021
Provenance baseline: 16/20 · Formally registered EU company (Sigmoid Creativity S.R.L.) with public registry numbers, SLA documentation, and developer SDKs.
Includes a 30-minute free trial without credit card; Pay As You Go credits start at $11 (valid 2 years); monthly subscriptions at $11/mo (10h), $28/mo (30h), and $90/mo (100h); annual subscriptions at $110/yr, $300/yr, and $900/yr; unused credits roll over up to 3x.
Pricing transparency baseline: 33/35 · Cancellation:
Audio files processed under EU GDPR and DPA; used strictly for deliverable rendering with full user ownership.
Privasee · Stripe
Product-level pricing & rules baseline
Verified plans, billing periods and cancellation terms owned by this product node; task-specific credit costs and free export limits are detailed per capability edge belowSitewide audio cleanup, voice enhancement, and podcast editing pricing · 2026-10-29
Full access to filler words removal, denoiser, studio sound, and timeline export；One-time trial quota；No auto-renewal；Until quota exhausted
Full feature access across web app and developer API；One-time credit purchase；No recurring renewal；Valid for 2 years
All features, multitrack editing, timeline export, credit rollover up to 30h
Rules [1] [2]All features, multitrack editing, timeline export, credit rollover；Billed annually at $110/year；Auto-renews annually；Valid for one year
All features, multitrack editing, timeline export, credit rollover up to 90h
Rules [1] [2]5 plan / billing-period records, all listed; — means that period is not recorded.
Includes a 30-minute free trial without credit card; Pay As You Go credits start at $11 (valid 2 years); monthly subscriptions at $11/mo (10h), $28/mo (30h), and $90/mo (100h); annual subscriptions at $110/yr, $300/yr, and $900/yr; unused credits roll over up to 3x.
Cancellation, sources & verification record
Cancellation and refund terms unverified
Cleanvoice AI Official Pricing Page ↗ · 2026-10-29 Capture 1 ↗
Archived official pricing sources: Cleanvoice AI Official Pricing Page (2026-10-29) [text snapshot ↗]
Capability map & linked task edges
Whenever a capability passes fixed-fixture testing, an edge is automatically linked between this product node and the task topic for horizontal comparison.
| Linked task edge | Score | Assessment · historical test | Test scope & delivery | Inspect capability / follow edge |
|---|---|---|---|---|
| Audio transcription | 70/100 | Cleanvoice AI includes 30 free signup minutes (no credit card) and its Summarize workflow generates speaker-labeled transcripts with native exports in 9 formats (.TXT, .SRT, .VTT, .TTML); it captures all 68 spoken words, though spaces around numeric punctuation and a ~280s async queue make it best suited for batch podcast post-production. | TXT · Output attached Historical test · 2026-11-05 Account / tier unconfirmed · 38.86s clean spoken English audio master (with verification token & numbers, 608 KB) · Output attached | |
| Voice isolation & denoising | 91/100 | Cleanvoice AI grants 30 minutes of free signup credit (no credit card; pay-as-you-go $11/5h or subscriptions from $11/mo) and its Clean Audio (Background noise) mode suppresses background noise to -96.8 dBFS (65/65 quality score) while natively exporting a watermark-free 321 kbps MP3 in ~71.6s. | MP3 · Output attached Historical test · 2026-11-05 Account / tier unconfirmed · 15s vocal & backing music mixed audio fixture (44.1kHz stereo, 353 KB) · Output attached |
Tested task edges (with scores and native artifacts) are strictly separated from discovered untested capabilities (unscored).
Peer product comparison across shared task edges
Matched by overlapping tested task edges; compare site-wide billing models alongside per-task benchmark scoresFree plan includes 10k credits/month (non-commercial); paid plans feature first-month promotions (Starter $1 first mo, then $5/mo; Creator $11 first mo, then $22/mo); overage billed per minute; taxes excluded.
- Voice isolation & denoising：Cleanvoice AI 91/100 vs ElevenLabs 97/100
Free tier provides 10 robotic voices and up to 1.5x speed; Premium tier costs $29/month or $139 billed yearly (approx. $11.58/mo, 60% savings) with 1000+ natural voices, 60+ languages, and voice typing.
- Audio transcription：Cleanvoice AI 70/100 vs Speechify 86/100
Free tier grants 3 transcriptions/day (max 30 min/file, no credit card needed); Unlimited tier offers unmetered transcription (up to 10 hrs or 5GB/file), priced at $120/year ($10/mo equivalent, 50% off) or $20 monthly; taxes excluded.
- Audio transcription：Cleanvoice AI 70/100 vs TurboScribe 72/100
Free tier provides 2 hours/month (includes Auphonic jingle, no rollover); annual recurring plans save 20% (Plan S $132/yr or $11/mo); one-time credits never expire ($12/5 hrs minimum); VAT/sales tax charged by region.
- Voice isolation & denoising：Cleanvoice AI 91/100 vs Auphonic 81/100
Linked capability test: Audio transcription
2026-11-05 · Native output attached. Findings cover the recorded fixture and conditions; paid specifications may be untested.
Verified native export & same-run screenshot evidence
Open native exported file (TXT · Output attached) ↗
Task quality
Word accuracy 79.4% · Keyword recall 60% · Verification token Fail · Audio duration 38.9s
Detailed facts & historical record
- Measured fixture results
Word accuracy: 79.4% · Keyword recall: 60% · Verification token: Fail · Audio duration: 38.9s
- Fixture findings
On the 38.86s English speech anchor, it captures all 68 words (79.4% word accuracy, 60.0% keyword recall), scoring 39.8/65 on strict string matching because punctuation formatting inserts spaces around numbers and hyphens (e.g., "$14 ,850 .50", "2026 -0814") and lemmatizes "Verified" to "Verify".
- Original historical observation
Uploaded the 38.86s speech master (anchor_transcribe_v1/audio.mp3) to the Cleanvoice Beta workspace (https://app.cleanvoice.ai/beta) with Summarize enabled; audio cleanup finished in 57.27s and queued transcription completed at 280.18s, exporting a 497 B native plain-text transcript (output.txt) via the Transcription Export dialog.
- Suitable uses · historical assessment
Not verified
- Result & output evidence
Delivery capabilities
TXT · Output attached · Check watermark record
Detailed facts & historical record
- Recorded export specifications
TXT · Output attached · Check watermark record
- Historical specification note · not an official maximum
Native .TXT plain text and .SRT / .VTT / .iTT / .TTML subtitle deliverables (68 words, 497 B)
- Watermark observation
Not verified
- Batch & API
Not verified
- Recorded free-tier dimensions
0.5
- Attached file dimensions
Not verified
Speed & reliability
About 281.68s end to end in this run; reliability not retested
Detailed facts & historical record
- Timing observation
Network-observed server wait: 280.18s · End-to-end reference: 281.68s
- Date & fixture
2026-11-05 · anchor_transcribe_v1 · US-West (California, AS2914)
- Repeated tests & reliability
No standardized repetition count, success rate or variability range is recorded.
Interaction experience
Account requirement unconfirmed
Detailed facts & historical record
- Account requirements
Not verified
- Sign-in methods
Supports Google OAuth, Microsoft OAuth, and email sign-in
- Manual adjustment
Not verified
- Interaction observations
Enabling Summarize (Creates transcript, summary, and social posts) after uploading audio automatically generates speaker-labeled (SPEAKER_00) timestamped transcripts and provides 9 native export options including Text Only, Text (Speakers & Timecodes), .SRT, .VTT, .iTT, and .TTML (02_result.png, 03_download.png).
Background & reputation
Sigmoid Creativity S.R.L. · Long-term reputation not independently verified
Detailed facts & historical record
- Service lifecycle notice
Not verified
- Operator
Sigmoid Creativity S.R.L.
- Recorded origin
Bucharest, Romania (founded in 2021)
- Domain history record
Domain registered in 2021
- Background & reputation assessment
Formally registered EU company (Sigmoid Creativity S.R.L.) with public registry numbers, SLA documentation, and developer SDKs.
- External reputation record
Operated by Romanian entity Sigmoid Creativity S.R.L. (Trade Registry J2021008970407, Tax RO44318031) with formal DPA, SLA, and EU GDPR privacy framework.
Engineering & privacy
Processing location unconfirmed；Explicit retention period unconfirmed; inspect policy record
Detailed facts & historical record
- Processing location
Processing location unconfirmed
- Third-party tracking record
Privasee · Stripe
- Engineering & privacy assessment
Operates under EU GDPR and formal DPA on European cloud clusters for acoustic enhancement and ASR transcription, with instant asset deletion in History.
- Retention statement
Explicit retention period unconfirmed; inspect policy record
Task free allowance & pricing disclosure
30 minutes of free processing upon registration (no credit card required); 38.86s audio clip deducts 1 minute credit
Not recorded
Site-wide plans belong to the product node (see Product-level pricing baseline ↑).
Expand site-wide plans aligned for Audio transcription (5 records)
Sitewide audio cleanup, voice enhancement, and podcast editing pricing · 2026-10-29
Full access to filler words removal, denoiser, studio sound, and timeline export；One-time trial quota；No auto-renewal；Until quota exhausted
Full feature access across web app and developer API；One-time credit purchase；No recurring renewal；Valid for 2 years
All features, multitrack editing, timeline export, credit rollover up to 30h
Rules [1] [2]All features, multitrack editing, timeline export, credit rollover；Billed annually at $110/year；Auto-renews annually；Valid for one year
All features, multitrack editing, timeline export, credit rollover up to 90h
Rules [1] [2]5 plan / billing-period records, all listed; — means that period is not recorded.
Includes a 30-minute free trial without credit card; Pay As You Go credits start at $11 (valid 2 years); monthly subscriptions at $11/mo (10h), $28/mo (30h), and $90/mo (100h); annual subscriptions at $110/yr, $300/yr, and $900/yr; unused credits roll over up to 3x.
Cancellation, sources & verification record
Cancellation and refund terms unverified
Cleanvoice AI Official Pricing Page ↗ · 2026-10-29 Capture 1 ↗
Evaluation basis & full scores · Overall 69.9/100
Overall = honesty & reputation × 30% + capability × 70%. Component maxima follow this report’s rubric. List scores are rounded for display; original precision is retained here.
Scoring version 2.3.1 · Test fixture anchor_transcribe_v1
Honesty & reputation · 91/100 · 30%
| Criterion | Score | Rationale | Basis |
|---|---|---|---|
| Pricing transparency | 33 / 35 | Credit and subscription tiers display exact per-hour rates, and the workspace header transparently displays remaining free trial minutes (Credit 29m -> 28m), deducting 1 credit for the 38.86s clip with no credit card required (01_landing.png). | Evidence A → |
| Interaction restraint | 28 / 30 | Enabling Summarize (Creates transcript, summary, and social posts) after uploading audio automatically generates speaker-labeled (SPEAKER_00) timestamped transcripts and provides 9 native export options including Text Only, Text (Speakers & Timecodes), .SRT, .VTT, .iTT, and .TTML (02_result.png, 03_download.png). | Evidence A → |
| Background & reputation | 16 / 20 | Formally registered EU company (Sigmoid Creativity S.R.L.) with public registry numbers, SLA documentation, and developer SDKs. | Evidence A → |
| Engineering & privacy | 14 / 15 | Operates under EU GDPR and formal DPA on European cloud clusters for acoustic enhancement and ASR transcription, with instant asset deletion in History. | Evidence A → |
Capability · 60.8/100 · 70%
| Criterion | Score | Rationale | Basis |
|---|---|---|---|
| Task quality | 39.8 / 65 | Standard 38.86s English speech master benchmark: achieved 79.4% word accuracy (68/68 words) and 60.0% keyword recall (6/10); points were deducted because punctuation formatting inserted spaces around hyphens and decimals (e.g. "$14 ,850 .50", "2026 -0814", "99 .4 %") and normalized "Verified" to "Verify" ("Realtest Transcribe Verify 2026") (39.8/65). | Evidence A → |
| Deliverable specifications | 16 / 20 | Free trial account natively exports plain text (.TXT), speaker/timecode text, .SRT, .VTT, .iTT, and Adobe Premiere .TTML subtitle files (16.0/20). | Evidence A → |
| Processing speed | 5 / 15 | Free-tier asynchronous queue first completed audio cleanup (57.27s) followed by queued ASR transcription and summarization in 280.18s total (5.0/15). | Evidence A → |
Legacy scores may include advertised maximum capabilities. Read tested output separately from paid capabilities. Records without matching tiers and scope are not ranked together.
Linked capability test: Voice isolation & denoising
2026-11-05 · Native output attached. Findings cover the recorded fixture and conditions; paid specifications may be untested.
Verified native export & same-run screenshot evidence
Open native exported file (MP3 · Output attached) ↗
Task quality
Vocal duration 15s · Duration retention 100% · Sample rate 44100 Hz · Channels Mono (1ch)
Detailed facts & historical record
- Measured fixture results
Vocal duration: 15s · Duration retention: 100% · Sample rate: 44100 Hz · Channels: Mono (1ch)
- Fixture findings
On the 15.0s 0 dB SNR cafe chatter and hum speech master, Cleanvoice AI achieves 100.0% background noise reduction (-96.8 dBFS residual noise in silent windows) and 100.0% duration retention (exact 15.0s with zero audio jingle), preserving crisp vocal clarity (1.662 speech energy ratio, 65.0/65).
- Original historical observation
Uploaded the 15.0s mixed audio master (anchor_voiceisolate_v1/mixed_audio.mp3) to the Cleanvoice Beta workspace (https://app.cleanvoice.ai/beta) and processed it with Clean Audio enabled (Keep Music off); cloud pipeline finished in 71.59s, deducting 1 minute of free credit (29m remaining) and exporting a 588.8 KB 44.1kHz 321kbps MP3 isolated vocal file (15.0s, 100.0% retention) plus DAW timeline marker exports.
- Suitable uses · historical assessment
Not verified
- Result & output evidence
Delivery capabilities
MP3 · Output attached · Check watermark record
Detailed facts & historical record
- Recorded export specifications
MP3 · Output attached · Check watermark record
- Historical specification note · not an official maximum
Not verified
- Watermark observation
Not verified
- Batch & API
Not verified
- Recorded free-tier dimensions
15 × 44100
- Attached file dimensions
Not verified
Speed & reliability
About 73.09s end to end in this run; reliability not retested
Detailed facts & historical record
- Timing observation
Network-observed server wait: 71.59s · End-to-end reference: 73.09s
- Date & fixture
2026-11-05 · anchor_voiceisolate_v1 · US-West (California, AS2914)
- Repeated tests & reliability
No standardized repetition count, success rate or variability range is recorded.
Interaction experience
Account requirement unconfirmed
Detailed facts & historical record
- Account requirements
Not verified
- Sign-in methods
Supports Google OAuth, Microsoft OAuth, and email sign-in
- Manual adjustment
Not verified
- Interaction observations
Once signed in, uploading mixed audio with Clean Audio enabled (and Keep Music unchecked) runs smoothly and provides an interactive dual-mode waveform player, DAW timeline marker exports (Audacity/Audition/Reaper), and direct 321kbps MP3 download with zero audio watermarks or paywall popups (02_result.png, 03_download.png).
Background & reputation
Sigmoid Creativity S.R.L. · Long-term reputation not independently verified
Detailed facts & historical record
- Service lifecycle notice
Not verified
- Operator
Sigmoid Creativity S.R.L.
- Recorded origin
Bucharest, Romania (founded in 2021)
- Domain history record
Domain registered in 2021
- Background & reputation assessment
Formally registered EU company (Sigmoid Creativity S.R.L.) with public registry numbers, SLA documentation, and developer SDKs.
- External reputation record
Operated by Romanian entity Sigmoid Creativity S.R.L. (Trade Registry J2021008970407, Tax RO44318031) with formal DPA, SLA, and EU GDPR privacy framework.
Engineering & privacy
Uploaded for cloud processing；Explicit retention period unconfirmed; inspect policy record
Detailed facts & historical record
- Processing location
Uploaded for cloud processing
- Third-party tracking record
Privasee · Stripe
- Engineering & privacy assessment
Operates under EU GDPR and formal DPA on European R2 storage and GPU clusters, with instant file management and deletion in History.
- Retention statement
Explicit retention period unconfirmed; inspect policy record
Task free allowance & pricing disclosure
30 minutes of free audio processing upon registration (no credit card required); 15s clip deducts 1 minute credit
Not recorded
Site-wide plans belong to the product node (see Product-level pricing baseline ↑).
Expand site-wide plans aligned for Voice isolation & denoising (5 records)
Sitewide audio cleanup, voice enhancement, and podcast editing pricing · 2026-10-29
Full access to filler words removal, denoiser, studio sound, and timeline export；One-time trial quota；No auto-renewal；Until quota exhausted
Full feature access across web app and developer API；One-time credit purchase；No recurring renewal；Valid for 2 years
All features, multitrack editing, timeline export, credit rollover up to 30h
Rules [1] [2]All features, multitrack editing, timeline export, credit rollover；Billed annually at $110/year；Auto-renews annually；Valid for one year
All features, multitrack editing, timeline export, credit rollover up to 90h
Rules [1] [2]5 plan / billing-period records, all listed; — means that period is not recorded.
Includes a 30-minute free trial without credit card; Pay As You Go credits start at $11 (valid 2 years); monthly subscriptions at $11/mo (10h), $28/mo (30h), and $90/mo (100h); annual subscriptions at $110/yr, $300/yr, and $900/yr; unused credits roll over up to 3x.
Cancellation, sources & verification record
Cancellation and refund terms unverified
Cleanvoice AI Official Pricing Page ↗ · 2026-10-29 Capture 1 ↗
Evaluation basis & full scores · Overall 91/100
Overall = honesty & reputation × 30% + capability × 70%. Component maxima follow this report’s rubric. List scores are rounded for display; original precision is retained here.
Scoring version 2.3.1 · Test fixture anchor_voiceisolate_v1
Honesty & reputation · 91/100 · 30%
| Criterion | Score | Rationale | Basis |
|---|---|---|---|
| Pricing transparency | 33 / 35 | Credit and subscription tiers display exact per-hour rates, and the workspace header transparently displays remaining free trial minutes (Credit 30m -> 29m), deducting 1 credit for the run with no credit card required (01_landing.png, 02_result.png). | Evidence A → |
| Interaction restraint | 28 / 30 | Once signed in, uploading mixed audio with Clean Audio enabled (and Keep Music unchecked) runs smoothly and provides an interactive dual-mode waveform player, DAW timeline marker exports (Audacity/Audition/Reaper), and direct 321kbps MP3 download with zero audio watermarks or paywall popups (02_result.png, 03_download.png). | Evidence A → |
| Background & reputation | 16 / 20 | Formally registered EU company (Sigmoid Creativity S.R.L.) with public registry numbers, SLA documentation, and developer SDKs. | Evidence A → |
| Engineering & privacy | 14 / 15 | Operates under EU GDPR and formal DPA on European R2 storage and GPU clusters, with instant file management and deletion in History. | Evidence A → |
Capability · 91/100 · 70%
| Criterion | Score | Rationale | Basis |
|---|---|---|---|
| Task quality | 65 / 65 | Deterministic acoustic & container check: cleanly isolated the 15.0s mixed audio master into a 15.0s high-fidelity vocal track (100.0% duration retention) at 44,100 Hz sample rate and 321 kbps bitrate (588.8 KB), stripping background music and noise with zero appended platform jingles (65.0/65). | Evidence A → |
| Deliverable specifications | 18 / 20 | Free trial account natively exports a 44.1kHz 321kbps high-bitrate MP3 vocal track (mixed_audio_clean.mp3, 588.8 KB) alongside Audacity / Audition / Reaper timeline markers, with zero watermarks or truncation (18.0/20). | Evidence A → |
| Processing speed | 8 / 15 | Free-tier asynchronous queue completed Prepare, Background noise, Leveling, and Export stages in 71.59s (8.0/15). | Evidence A → |
Legacy scores may include advertised maximum capabilities. Read tested output separately from paid capabilities. Records without matching tiers and scope are not ranked together.
Developer & citation kit · Claim updates, request benchmarks & embed verified badges
Product developers and writers can embed these SVG badges in official websites, GitHub READMEs or documentation, or use the channels below to sync product-wide plans and capabilities.
[](https://realtest.org/tools/cleanvoice.ai/) <a href="https://realtest.org/tools/cleanvoice.ai/"><img src="https://realtest.org/badges/cleanvoice.ai.svg" alt="Cleanvoice AI — RealTest" height="24" /></a> 
[](https://realtest.org/tools/cleanvoice.ai/#feature-audio-transcription) 
[](https://realtest.org/tools/cleanvoice.ai/#feature-voice-isolator)