Perso Linnet4
Positive judgments across voice, naturalness and emotion in two scenes.
generated: 3/3 · held judgments: 0Listen to Perso Linnet4 and ElevenLabs on the same three films. Compare Spanish dubbing, inspect the AI observations, then choose the workflow that fits your video.

This is a Perso-authored pilot, not an independent certification or a statistically established winner. AI labels describe these samples only. Held judgments and failed outputs remain visible.
Positive judgments across voice, naturalness and emotion in two scenes.
generated: 3/3 · held judgments: 0Strong findings on Sintel; emotion preservation stood out on Tears of Steel.
generated: 3/3 · held judgments: 2Compare the original with each first output. Playing one sample pauses the others. Open a film to explore its samples and measurements.

| AI observations, by scene | Perso Linnet4 | ElevenLabs |
|---|---|---|
| Translation | Meaning preserved in cited examples | Meaning preserved in cited examples |
| Voice similarity | Strong similarity | Strong similarity |
| Naturalness | No clear artifacts detected | No clear artifacts detected |
| Emotion | Emotion preserved | Emotion preserved |
| Scene-level automatic measurements | Role 1 / role 2 cosine | UTMOSv2 predicted mean |
|---|---|---|
| Perso Linnet 4 | 0.408 / 0.302 | 2.61 |
| ElevenLabs Dubbing v2 | 0.470 / 0.270 | 2.23 |

| AI observations, by scene | Perso Linnet4 | ElevenLabs |
|---|---|---|
| Translation | Meaning preserved in cited examples | Meaning preserved in cited examples |
| Voice similarity | Partial similarity | Judgment withheld |
| Naturalness | Minor artifacts | Minor artifacts |
| Emotion | Partial emotion shift | Emotion preserved |
| Scene-level automatic measurements | Role 1 / role 2 cosine | UTMOSv2 predicted mean |
|---|---|---|
| Perso Linnet 4 | 0.432 / 0.425 | 2.14 |
| ElevenLabs Dubbing v2 | 0.492 / 0.365 | 2.18 |

| AI observations, by scene | Perso Linnet4 | ElevenLabs |
|---|---|---|
| Translation | Meaning preserved in cited examples | Meaning preserved in cited examples |
| Voice similarity | Strong similarity | Judgment withheld |
| Naturalness | No clear artifacts detected | Minor artifacts |
| Emotion | Emotion preserved | Partial emotion shift |
| Scene-level automatic measurements | Role 1 / role 2 cosine | UTMOSv2 predicted mean |
|---|---|---|
| Perso Linnet 4 | 0.589 / 0.255 † | 2.54 |
| ElevenLabs Dubbing v2 | 0.563 / 0.449 | 2.34 |
Cosine is not a probability; UTMOSv2 is not human MOS. Background audio and short segments affect these diagnostics. They are not combined into ranking scores. Speaker boundaries may be uncertain.
We distinguish recognized evaluation frameworks, reproducible automatic measurements and AI interpretation. These results are not an independent certification.
Perso · Sintel: the AI explanation refers ambiguously to speakers. Its positive voice-similarity label is retained with this warning, not treated as verified speaker identity.
Perso · Cosmos: role 2 cosine is 0.255 against the original role and 0.244 against the other role. Interpretation is withheld; this cannot establish voice preservation or a speaker swap.
How we evaluated ↗The product and pricing information below concerns available workflows and plans. It is not the cost or quality score of the Linnet4 sample test. Check current rates, model access, lip-sync and revision limits before purchasing.
Evaluated November 18, 2026 · Perso-authored pilot
Tears of Steel / Sintel: Blender Foundation, CC BY 3.0. Cosmos Laundromat: Blender Foundation, CC BY 4.0.
Sintel · Tears of Steel · Cosmos Laundromat
Compare the specific product and plan. On mobile, swipe the table sideways to see both tools.
| Feature | Perso Dubbing | ElevenLabs |
|---|---|---|
| Product compared | AI Dubbing | Dubbing v2 |
| Existing video input | Yes | Yes; translates the audio |
| Source-video lip-sync | Lip-dubbing supported | Not part of Dubbing; available elsewhere via third-party models |
| Multiple source voices | Speaker detection and voice preservation | Speaker detection and voice preservation |
| Edit a generated dub | Script review; verify regeneration usage | No in-app editor; Enterprise API editing |
| Output | Video and supported audio/script downloads | Single lossless audio file |
| Legacy product distinction | Check the selected output mode | Dubbing Studio v1 offers editing and multiple export formats |
Dubbing v2 provides automatic translation with speaker detection and returns a single lossless audio file. It has no in-app transcript editor; Enterprise plans support transcript editing and regeneration through the API. Legacy Dubbing Studio uses v1, remains in maintenance mode and supports MP4, AAC, AAF, SRT and separate speaker WAV exports.
Choose Perso Dubbing when you want to upload existing footage, review the script and create a dubbed video with lip-dubbing in one workflow. Confirm your language pair, clip length and download requirements on a test project.
Choose ElevenLabs when translated audio is the main deliverable or your production already uses its voice tools and APIs. If you need detailed editing, distinguish legacy Dubbing Studio from v2 and verify Enterprise API access.
Perso Dubbing offers Finch dubbing from $0.11/min on PRO monthly: $99/month for 10,800 credits, or up to 900 minutes at full Finch allocation. Actual minutes depend on the model and other credit use. Optional lip sync has separate usage conditions. ElevenLabs charges dubbing by source duration and target language, with rates depending on the model. Include any additional video assembly or lip-sync workflow in a like-for-like estimate.
Use source duration × target languages, then account for lip-sync, revisions and plan limits. Monthly billing and annual promotions are different commitments. Prices and allowances can change.
Lip-sync is not part of ElevenLabs Dubbing. ElevenLabs documents lip-sync in other products through third-party models. This comparison evaluates Dubbing v2, not every tool on the platform.
Legacy Dubbing Studio supports separate speaker WAV tracks and formats including MP4 and SRT. Dubbing v2 returns a single lossless audio file. Check which model your workflow uses.
Perso offers a free trial. Check the current pricing page for trial credits, output limits and access to lip-dubbing before starting your test.
Use the same source clip, language pair and output settings. Review names, translation meaning, voice consistency and visible lip movements. Include more than one speaker when that matches your real workload.
Start with your original source video and approved transcript. Upload the source, choose a supported target language, review the generated script and create a test output. Check file limits and export requirements before moving a full library.
Prepared by Perso Dubbing. Product information was checked against the official pages below on September 14, 2026. Confirm current availability and usage rules before purchasing.