# ElevenLabs vs Resemble AI (2026): Is Resemble still an option for text to speech?

By Samuel Oram-Jones

Resemble AI's chief executive wrote on 13 August 2026 that "We're not selling voice AI to new customers", which settles most of the ElevenLabs vs Resemble AI question before price comes up. Resemble says it supports its existing voice customers, takes no new ones, and has put the whole company behind deepfake detection.

A team shopping for a text to speech API is therefore choosing between ElevenLabs and everything else. Bland Speech is one of the everything else: it costs $40 per million characters and bills pay as you go, and Bland Speech v3 ranked first of 13 AI voices in Design Arena's Audio Realism Bench results of 4 August 2026, where ElevenLabs' Eleven v3 ranked 12th.

## Which is better, ElevenLabs or Resemble AI?

ElevenLabs is the better pick for a team buying text to speech today, because ElevenLabs sells it and Resemble AI doesn't sell it to new customers, as of November 2026. Resemble says its generative voice models "are no longer offered as commercial paid products", and Resemble now sells detection of AI-generated audio, video, and images.

| Feature | Bland Speech | ElevenLabs | Resemble AI |
|---|---|---|---|
| Sold to new customers | Yes, pay as you go | Yes, Free to Enterprise plans | No, for voice AI since August 2026 |
| Audio Realism Bench, 4 August 2026 | Ranked 1st of 13, 1365 | Eleven v3: ranked 12th, 938 | Wasn't in those results |
| Price per million characters | $40, one standard rate | $80 Eleven v4, $40 Eleven v4 Turbo | No text to speech price, plans cover deepfake detection |
| Instant voice clone | About 10 seconds of audio | 10 seconds, from the Starter plan | Not sold to new customers |
| What the company sells now | A text to speech API and studio | Text to speech models on plans from Free to Enterprise | Deepfake detection and watermarking |

<details><summary>What these rows mean</summary>

- **Sold to new customers:** Whether a team that signs up today can buy the product. Resemble says it supports existing customers and takes no new ones.
- **Audio Realism Bench, 4 August 2026:** Design Arena's blind listening test, where listeners hear two clips of the same line and pick the one that sounds more like a real person. Resemble AI wasn't in those results, and ElevenLabs' newer Eleven v4 models were released after them.
- **Price per million characters:** ElevenLabs charges $0.08 per 1,000 characters for Eleven v4 and $0.04 for Eleven v4 Turbo on its API. Bland Speech charges $0.04 per 1,000.
- **Instant voice clone:** A clone made from one short recording, with no training run.

</details>

## What are the catches with ElevenLabs and Resemble AI?

ElevenLabs' catch is cost, and Resemble AI's catch is that no voice product is on sale, only open models a team hosts itself.

### ElevenLabs

ElevenLabs takes new customers on every plan from Free to Enterprise. Eleven v4 and Eleven v4 Turbo launched on 28 September 2026, Eleven v4 speaks 90+ languages, and ElevenLabs says Eleven v4 Turbo has a median latency of about 100ms.

Cloning has two tiers. An instant clone takes 10 seconds of audio on Eleven v4 from the Starter plan, and a professional clone takes about 30 minutes of audio from the Creator plan up, with the voice owner reading a captcha aloud to confirm the voice is theirs. ElevenLabs holds SOC 2 Type 2 and PCI DSS 4.0.1 Level 1, names HIPAA in its trust center, and says enterprises can run text to speech inside their own cloud infrastructure.

<div class="cmp-proscons" data-rival="elevenlabs"><div class="cmp-proscons-pros"><p><strong>Pros</strong></p><ul><li>A hosted text to speech API that new customers can buy today</li><li>90+ languages across the Eleven v4 family</li><li>Instant and professional clones, with voice-captcha verification</li><li>Mu-law and A-law output at 8 kHz for phone lines</li></ul></div><div class="cmp-proscons-cons"><p><strong>Cons</strong></p><ul><li>Eleven v3 ranked 12th of 13 AI voices on Audio Realism Bench on 4 August 2026</li><li>$80 per million characters for Eleven v4 on the API</li><li>A Free plan for non-commercial use only</li><li>Plan allowances to track, from 10,000 characters on Free to 12,375,000 on Business</li></ul></div></div>

### Resemble AI

Resemble AI built text to speech and voice cloning products and then stopped selling them. Resemble says "The voice models we built are open and free to use, but we won't be producing new ones for commercial sale", and the company now sells detection of AI-generated media, with plans at $0 on Flex, $350 a month on Team, $1,000 a month on Business, and Enterprise custom, all priced for detection and verification work.

What is left is Chatterbox, a family of open text to speech models under the MIT license. Chatterbox Multilingual V3 has 500 million parameters and covers 23+ languages, Chatterbox-Turbo is an English model of 350 million parameters made for low-latency voice agents, and every output carries Resemble's PerTh watermark. Resemble says "As of August 2026, Chatterbox is retired research and is no longer maintained", with no support, updates, or paid tier. A team that uses it hosts the model, runs the hardware, and owns every fix.

<div class="cmp-proscons" data-rival="resemble-ai"><div class="cmp-proscons-pros"><p><strong>Pros</strong></p><ul><li>Chatterbox weights are free under the MIT license</li><li>Every Chatterbox output carries a PerTh watermark</li><li>Chatterbox Multilingual V3 covers 23+ languages</li><li>A deepfake detection product for audio, video, and images</li></ul></div><div class="cmp-proscons-cons"><p><strong>Cons</strong></p><ul><li>No text to speech or voice cloning sold to new customers</li><li>Chatterbox is retired and no longer maintained</li><li>No support, updates, or paid tier for the open models</li><li>A team hosts the model itself</li></ul></div></div>

## What do ElevenLabs, Resemble AI, and Bland Speech cost?

Bland Speech costs $40 per million characters with no monthly plan, ElevenLabs costs $40 to $80 per million depending on the model and plan, and Resemble AI has no text to speech price to quote.

### ElevenLabs pricing

ElevenLabs charges $0.08 per 1,000 characters for Eleven v4 and $0.04 for Eleven v4 Turbo on its API, with plans at $6, $22, $99, $299, and $990 a month and a custom Enterprise tier. At the Eleven v4 rate the paid plans include between 75,000 and 12,375,000 characters, and the Free plan, for non-commercial use, includes 10,000.

### Resemble AI pricing

Resemble's plans cover deepfake detection, intelligence, identity, watermarking, and agent detection, with no text to speech line. The Chatterbox models cost nothing to download, with compute on the user's own bill.

### Bland Speech pricing

Bland Speech costs $0.04 per 1,000 characters, directed or not, in the studio and the API, pay as you go with no plan. A new account gets 50,000 characters free, once. Enterprise pricing is custom, set with Bland's sales team.

### What a month of speech costs on each

At 100,000 characters a month, Bland Speech costs $4 and ElevenLabs costs $22 on Creator for Eleven v4 or $6 on Starter for Eleven v4 Turbo. At 1 million characters, Bland Speech costs $40 and ElevenLabs costs $99 on Pro. At 10 million, Bland Speech costs $400 and ElevenLabs costs $990 on Business, assuming usage stays inside the plan's included characters.

## Which sounds more human, ElevenLabs, Resemble AI, or Bland Speech?

Resemble AI wasn't in Design Arena's Audio Realism Bench results of 4 August 2026, so those results hold no Resemble score. In those results Bland Speech v3 ranked first of 13 AI voices at 1365, 427 points above ElevenLabs' Eleven v3 in 12th at 938, and a human recording scored 1500.

Listeners hear two clips of the same line without knowing which model made which, and pick the one that sounds more like a real person. Scores are Elo ratings, the chess-style number that rises every time a voice is picked over another. ElevenLabs released Eleven v4 and Eleven v4 Turbo on 28 September 2026, after those results, and neither is scored in them. Design Arena set out the method and the 4 August scores in [its Audio Realism Bench article](https://notes.designarena.ai/audio-realism-benchmark-measuring-realism-in-audio-models/).

## Why do people comparing ElevenLabs and Resemble AI choose Bland Speech?

Bland Speech is a hosted alternative to ElevenLabs for a team that Resemble AI won't take on, with Bland Speech v3 ranked first of 13 AI voices on 4 August 2026 and one standard rate. A team that came to Resemble for cloning can start with an instant clone from one clean sample of about 10 seconds, or a professional clone fine-tuned on 30 minutes or more of audio after Bland's consent and rights review. Every clone is private to the customer's organization, and the customer confirms the voice is theirs to use.

A voice cloned on one provider's service doesn't move to another, so a team leaving a Resemble voice re-clones from the original recordings, with the speaker's consent. Bland Speech streams over HTTP and a WebSocket at /v2/tts/ws, returns mu-law at 8 kHz or PCM from 8 to 48 kHz, and publishes 199ms to first audio, which is slower than the 100ms ElevenLabs publishes for Eleven v4 Turbo. The [text to speech API](/text-to-speech-api) page has the endpoints.

Bland is SOC 2 Type II audited, HIPAA compliant with a signed BAA, and FedRAMP 20x Class C certified at Moderate, granted on 17 November 2026. Unlike Chatterbox's open weights, Bland Speech is a closed-source model Bland built itself.

<div class="cmp-proscons cmp-proscons-own"><div class="cmp-proscons-pros"><p><strong>Pros</strong></p><ul><li>Bland Speech v3 ranked first of 13 AI voices on Audio Realism Bench on 4 August 2026</li><li>One standard rate, $0.04 per 1,000 characters, with no monthly plan</li><li>Instant clones from about 10 seconds, professional clones after a consent and rights review</li><li>Mu-law and PCM output ready for phone calls</li></ul></div><div class="cmp-proscons-cons"><p><strong>Cons</strong></p><ul><li>More than a hobby project needs, because Bland Speech is built for production calls</li></ul></div></div>

Bland has pages on [Bland Speech against Resemble AI](/compare/bland-speech-vs-resemble-ai), a [Resemble AI review](/blog/resemble-ai-review), and a list of [Resemble AI alternatives](/blog/resemble-ai-alternatives). The [Bland Speech](/speech) page has the studio and the price, the [text to speech comparison hub](/compare/text-to-speech) covers every vendor, and the [rival comparisons](/compare/text-to-speech/rival-vs-rival) cover other pairs.

## Frequently asked questions

### Does Resemble AI still sell text to speech?

No. Resemble says it is "not selling voice AI to new customers", supports existing customers, and has moved the company to deepfake detection. Resemble also says it does not sell text-to-speech, voice cloning, or any other voice AI product.

### Is ElevenLabs better than Resemble AI?

For a team buying text to speech now, yes, because ElevenLabs sells a hosted API with instant and professional voice clones and Resemble AI doesn't sell one to new customers. Bland Speech is the other hosted option, at $40 per million characters with no monthly plan.

### Is Resemble AI's Chatterbox free?

Chatterbox is free to download and run under the MIT license, and Resemble calls it retired research with no support, updates, or paid tier. A team that uses it hosts and supports the model itself.

### Can a new customer still clone a voice with Resemble AI?

Resemble says it is not taking new voice customers and that its open models are free but won't be produced for commercial sale. It still sells identity enrollment for detection and matching, which checks whether a voice is a particular person's and doesn't generate speech.

### What does Resemble AI sell now?

Resemble sells multimodal deepfake detection and watermarking for enterprises. Its plans are Flex at $0, Team at $350 a month, Business at $1,000 a month, and Enterprise.

### Which text to speech API suits a voice agent on a phone line?

Not Resemble AI, which no longer sells speech to new customers. Of the providers still selling, Bland Speech v3 ranked first of 13 AI voices in the Audio Realism Bench results of 4 August 2026 and Bland Speech returns mu-law at 8 kHz, while ElevenLabs offers 90+ languages and voice-captcha checks on professional clones.
