elevenlabs.io · Voice isolation & denoising
ElevenLabs Voice Isolator demonstrates studio-grade neural voice isolation. On the 15.0s benchmark audio, server processing took just 8.36s, yielding pristine 44.1kHz 320kbps MP3 audio with 99.9% duration retention and complete noise removal. Free tiers receive 10,000 monthly credits, making it an outstanding utility.
Tested export MP3 · Output attached
Historical test · 2026-11-04
Account / tier unconfirmed · 15s vocal & backing music mixed audio fixture (44.1kHz stereo, 353 KB) · Output attached
Task quality
Vocal duration 14.98s · Duration retention 99.9% · Sample rate 44100 Hz · Channels Mono (1ch)
Detailed facts & historical record
- Measured fixture results
Vocal duration: 14.98s · Duration retention: 99.9% · Sample rate: 44100 Hz · Channels: Mono (1ch)
- Fixture findings
8.36s isolation yielding 44.1kHz 320kbps pristine vocals, 99.9% duration retention, and unwatermarked native MP3 export.
- Original historical observation
Uploaded 15.0s benchmark audio (anchor_voiceisolate_v1) to ElevenLabs Voice Isolator; cloud neural engine finished in 8.36s consuming 251 credits, yielding studio-grade 44.1kHz 320kbps MP3 with 99.9% duration retention and flawless noise removal.
- 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
44,100 Hz 320kbps native MP3 high-fidelity vocal export
- Watermark observation
Not verified
- Batch & API
Not verified
- Recorded free-tier dimensions
14.98 × 44100
- Attached file dimensions
Not verified
Speed & reliability
About 9.56s end to end in this run; reliability not retested
Detailed facts & historical record
- Timing observation
Network-observed server wait: 8.36s · End-to-end reference: 9.56s
- Date & fixture
2026-11-04 · 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
Google login or email login
- Manual adjustment
Not verified
- Interaction observations
Clean studio workspace with immediate one-click download of native MP3; no popups or coercive upgrade gates.
Background & reputation
ElevenLabs Inc. · Long-term reputation not independently verified
Detailed facts & historical record
- Service lifecycle notice
Not verified
- Operator
ElevenLabs Inc.
- Recorded origin
United States
- Domain history record
Founded in 2022
- Background & reputation assessment
Delaware corporation founded by former Google and Palantir engineers; well recognized in generative audio.
- External reputation record
Industry-leading neural speech synthesis and generative audio platform.
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
Google Analytics · Datadog
- Engineering & privacy assessment
Enterprise cloud audio processing with HTTPS encryption, zero training on non-commercial user audio.
- Retention statement
Explicit retention period unconfirmed; inspect policy record
Pricing & rules
ElevenLabs sitewide audio generation, voice cloning, and transcription credits; monthly and free tiers · 2026-10-29
4 plan / billing-period records, all listed; — means that period is not recorded.
Free 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.
Cancellation, sources & verification record
Cancellation and refund terms unverified
ElevenLabs official homepage ↗ · 2026-10-29 Capture 1 ↗
ElevenLabs official pricing plans page ↗ · 2026-10-29 Capture 1 ↗
Allowance & disclosure on test date
Not recorded
Not recorded
Evaluation basis & full scores · Overall 96.7/100
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 · 96/100 · 30%
| Criterion | Score | Rationale | Basis |
|---|---|---|---|
| Pricing transparency | 35 / 35 | Transparent credit usage and quota display; free tier provides 10,000 credits/mo with clear usage indicators. | Evidence A → |
| Interaction restraint | 28 / 30 | Clean studio workspace with immediate one-click download of native MP3; no popups or coercive upgrade gates. | Evidence A → |
| Background & reputation | 19 / 20 | Delaware corporation founded by former Google and Palantir engineers; well recognized in generative audio. | Evidence A → |
| Engineering & privacy | 14 / 15 | Enterprise cloud audio processing with HTTPS encryption, zero training on non-commercial user audio. | Evidence A → |
Capability · 97/100 · 70%
| Criterion | Score | Rationale | Basis |
|---|---|---|---|
| Task quality | 65 / 65 | Isolated speech from 15.0s benchmark audio in 8.36s, exporting pristine 44.1kHz 320kbps MP3 with 99.9% duration retention and complete noise removal. | Evidence A → |
| Deliverable specifications | 18 / 20 | Exports native 44.1kHz 320kbps MP3 deliverable (587.8 KB) with full acoustic fidelity and zero watermarks. | Evidence A → |
| Processing speed | 14 / 15 | Cloud neural audio pipeline finished processing in 8.36s. | 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.