# Bland Speech vs Amazon Polly (2026): Is a cheaper voice worth it on a phone call?

Amazon Polly charges $4 per million characters for Standard voices and $30 for Generative ones, so a team can pay a tenth of Bland Speech's $40 or three quarters of it depending on the engine it picks. Polly also includes 5 million Standard characters free every month, so a team under that volume pays nothing for Standard voices.

The price is the reason to look at Polly, and the voice is the open question. Polly isn't in Design Arena's realism results, Polly offers cloning as a Brand Voice engagement through an AWS account manager, and the strongest case for Bland Speech rests on what a caller hears in the first second.

## Which is better, Bland Speech or Amazon Polly?

For phone calls and voice agents where callers should hear a person, Bland Speech is the better choice as of November 2026. Bland Speech v3 ranked first of 13 AI voices in Design Arena's Audio Realism Bench results of 4 August 2026, at 1365, while Amazon Polly wasn't in those results, so they give no score to set beside it.

Amazon Polly makes more sense for a team that already runs on AWS and wants the lowest bill for alerts, notifications, and IVR prompts, where a voice only has to be clear. A team leaving Polly can read the [migration guide](/blog/migrate-from-amazon-polly), the [Amazon Polly alternatives](/blog/amazon-polly-alternatives) list, or the [Polly review](/blog/amazon-polly-review).

<div class="cmp-usewhen"><div class="cmp-usewhen-own"><p><strong>Use Bland Speech when</strong></p><ul><li>Callers on a phone line should hear a voice that sounds like a person</li><li>A team wants to write a laugh, a sigh, or a warm read into a script with tags such as [laughs] or [say warmly]</li><li>A team wants an instant voice clone from about 10 seconds of audio, with no account manager in the loop</li><li>Text arrives from a language model in pieces and should stream into a WebSocket that supports barge-in</li></ul></div><div><p><strong>Use Amazon Polly when</strong></p><ul><li>A team wants the lowest per-character price, with Standard voices at $4 per million and 5 million characters free each month</li><li>A team already runs on AWS and wants speech billed and governed there, under the AWS BAA for HIPAA workloads</li><li>A script needs SSML controls for pauses, pace, volume, emphasis, phonetic pronunciation, breathing sounds, and whispering</li><li>A team wants cached speech replayed at no additional cost</li><li>A phone integration needs A-law as well as mu-law at 8 kHz</li></ul></div></div>

| Feature | Bland Speech | Amazon Polly |
| --- | --- | --- |
| Audio Realism Bench, 4 August 2026 | #1 of 13 AI voices, 1365 (Bland Speech v3) | Not in the results |
| Price per million characters | $40, one rate | $4 Standard, $16 Neural, $30 Generative, $100 Long-Form |
| Monthly plan | None, pay as you go | None, pay per character |
| Free usage | 50,000 characters, once | 5 million Standard characters a month, with smaller tiers for other engines |
| Text per request over HTTP | Up to 5,000 characters | Up to 3,000 billed characters |
| Phone audio | Mu-law at 8 kHz, PCM from 8 to 48 kHz | Mu-law and A-law at 8 kHz, PCM at 8 and 16 kHz |
| Streaming | HTTP chunked and a WebSocket | HTTP streaming, and bidirectional streaming over HTTP/2 on Generative |
| Voice cloning | Instant from about 10 seconds, professional from 30 minutes or more | Brand Voice through an AWS account manager |

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

- **Audio Realism Bench, 4 August 2026:** Design Arena's blind listening test, where native speakers hear two recordings of the same line and pick the one that sounds more like a real person. The 13 AI voices in the results came from other companies, so Amazon Polly has no score.
- **Price per million characters:** Bland Speech costs $0.04 per 1,000 characters, directed or not. Amazon Polly charges $4.00 for Standard, $16.00 for Neural, $100.00 for Long-Form, and $30 for Generative per million characters, outside its free tier.
- **Monthly plan:** Polly bills per character with no plan tiers on its pricing, and Bland Speech has no subscription.
- **Free usage:** New Bland accounts get 50,000 characters free. Polly's free tier is 5 million Standard characters a month, and 1 million Neural, 500 thousand Long-Form, and 100 thousand Generative characters a month for the first 12 months. New AWS customers also get up to $200 in credits.
- **Text per request:** Bland's HTTP endpoint takes up to 5,000 characters, and Polly's SynthesizeSpeech takes up to 3,000 billed characters, or 6,000 total.
- **Phone audio:** Mu-law is the 8-bit format US phone networks and Twilio carry, and A-law is its European counterpart.
- **Streaming:** A WebSocket is a connection that stays open so text and audio flow both ways.

</details>

## How do Bland Speech and Amazon Polly differ?

Amazon Polly is a pay-per-character service with four voice engines at four prices, while Bland Speech is one model at one price built for live phone calls. The engines in Polly range from Standard to Generative, which AWS calls its largest Polly model to date.

### Bland Speech overview

Bland Speech is Bland's text to speech model, sold on its own as an API and a studio on the [Bland Speech](/speech) page. Bland is a voice AI provider and runs its own model and speech.

Text goes to POST /v2/tts over HTTP and comes back as raw audio or a WAV file, or streams into WSS /v2/tts/ws, a WebSocket where a new context interrupts the turn that is playing, so a caller can cut in. That is called barge-in. Performance tags in square brackets tell the model how to deliver the words that follow. The [text to speech API](/text-to-speech-api) page has the endpoints and code.

<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 rate, $0.04 per 1,000 characters, in the studio and the API</li><li>HTTP and WebSocket streaming, with barge-in in the protocol</li><li>Performance tags for how a line is read</li><li>Mu-law at 8 kHz and PCM from 8 to 48 kHz</li><li>Instant clones from about 10 seconds of audio, private to the customer's organization</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>

### Amazon Polly overview

Amazon Polly offers four voice engines: Standard, Neural, Long-Form, and Generative. AWS describes Generative voices as emotionally engaged, assertive, and highly colloquial, and says Polly offers 55 voices in a generative variant. Control comes from SSML, the markup language that tells a speech engine where to pause or what to stress, in a subset of the W3C 1.1 specification.

Amazon Polly streams audio over HTTP, and its StartSpeechSynthesisStream call takes text and returns audio on one bidirectional HTTP/2 connection on the Generative engine in listed regions.

<div class="cmp-proscons" data-rival="amazon-polly"><div class="cmp-proscons-pros"><p><strong>Pros</strong></p><ul><li>Standard voices at $4 per million characters, with 5 million free each month</li><li>Cached speech replays at no additional cost</li><li>Mu-law and A-law output at 8 kHz</li><li>SSML pauses, pace, volume, emphasis, pronunciation, breathing sounds, and whispering</li><li>HIPAA Eligible Service under the AWS BAA, and in scope for SOC 1, 2, and 3</li></ul></div><div class="cmp-proscons-cons"><p><strong>Cons</strong></p><ul><li>Wasn't one of the voices tested in the Audio Realism Bench results of 4 August 2026</li><li>Cloning is a Brand Voice engagement through an AWS account manager</li><li>Newscaster style and speech marks aren't available on the Generative engine</li><li>AWS says generative voices can vary slightly between model updates</li><li>Long-Form voices cost $100 per million characters</li></ul></div></div>

### How do the two handle phone audio and streaming?

Bland publishes 199ms to first audio on a streamed request, and time to first audio is the gap between sending text and hearing the first sound. Both services stream: Bland over HTTP chunks and a WebSocket, and Polly over HTTP, with bidirectional streaming on the Generative engine.

On formats, Bland Speech returns mu-law at 8 kHz or PCM at 8,000 to 48,000 Hz. Amazon Polly returns mp3, ogg_vorbis, ogg_opus, mu-law, a-law, and pcm, at 8,000 Hz for mu-law and a-law and 8,000 or 16,000 Hz for pcm.

### How do the two handle voice cloning?

Bland Speech makes an instant clone from one clean sample of about 10 seconds, and a professional clone is 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.

Amazon Polly offers cloning as a Brand Voice, and AWS tells teams interested in one to reach out to their AWS account manager.

### How do the two handle security?

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, and Bland's [trust page](/trust-security) says speech synthesis runs inside the FedRAMP boundary. Enterprise customers get dedicated orchestration and dedicated GPUs run by Bland.

AWS says Amazon Polly is a HIPAA Eligible Service covered under the AWS BAA, and AWS says Polly is certified for use with regulated workloads for HIPAA and PCI DSS. Polly is in scope for SOC 1, 2, and 3.

## How much do Bland Speech and Amazon Polly cost?

Amazon Polly costs less per character than Bland Speech on every engine except Long-Form: $4 Standard, $16 Neural, and $30 Generative against $40, with Long-Form at $100. Bland Speech bills one rate in the studio and the API.

### Bland Speech pricing

Bland Speech costs $0.04 per 1,000 characters, directed or not, in the studio or the API, billed against a balance that doesn't expire. New accounts get 50,000 characters free. Bland's own examples are a 60-second note at $0.04, a 10-minute narration at $0.36, and a million characters at $40.

Enterprise pricing is custom, set with Bland's sales team.

### Amazon Polly pricing

Amazon Polly charges per character converted to speech or to Speech Marks metadata, with no plan tiers: $4.00 per million for Standard, $16.00 for Neural, $100.00 for Long-Form, and $30 for Generative, after the free tier. The free tier is 5 million Standard characters a month, with no time limit stated, and 1 million Neural, 500 thousand Long-Form, and 100 thousand Generative characters a month for the first 12 months.

### What does a month of speech cost on each?

Before free tiers, 100,000 characters costs $4 on Bland Speech and $0.40 on Polly Standard, $1.60 on Neural, and $3 on Generative. A million characters is $40 against $4, $16, and $30, and ten million is $400 against $40, $160, and $300.

Polly's free tier takes Standard use under 5 million characters a month to zero. What the extra spend on Bland Speech buys is a voice built for live calls, with performance tags and instant clones, and for a delivery alert none of that is the point, so Polly Standard is the sensible buy there.

## Which sounds more human, Bland Speech or Amazon Polly?

Amazon Polly wasn't one of the voices in Design Arena's Audio Realism Bench results of 4 August 2026, where Bland Speech v3 ranked first of 13 AI voices at 1365 and the human recording scored 1500, so the results can't say which of the two sounds more human.

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 system chess uses, where a model gains more for beating a higher-rated one. Polly's own description of its Generative engine is that it is emotionally engaged, assertive, and highly colloquial, and AWS notes that generative voices can vary slightly when it updates the model, so a team would test the voice it ships on. Design Arena's [published results](https://notes.designarena.ai/audio-realism-benchmark-measuring-realism-in-audio-models/) explain how the votes were collected.

## Why is Bland Speech the best Amazon Polly alternative?

Bland Speech is the better Amazon Polly alternative when a caller's first impression of the voice matters more than the price per character. Bland Speech v3 ranked first among AI voices on Audio Realism Bench on 4 August 2026, an instant clone takes about 10 seconds of audio, and delivery can be written into a script with tags.

The [migration guide](/blog/migrate-from-amazon-polly) maps Polly concepts to Bland's, and cloned voices need re-cloning from the original recordings with consent, because they aren't portable between providers. Bland's [comparison of text to speech APIs](/compare/text-to-speech) covers more vendors, and the [head-to-heads](/compare/text-to-speech/head-to-heads) hub lists every pairing.

### How was this comparison made?

Amazon Polly details were captured from AWS's Polly pricing, FAQ, developer guide, API reference, quotas, and compliance pages, and voice scores come from Design Arena's published results of 4 August 2026.

## Frequently asked questions

### Is Bland Speech better than Amazon Polly?

For phone calls and voice agents, Bland Speech is the better choice. Bland Speech v3 ranked first of 13 AI voices on Audio Realism Bench on 4 August 2026, and Polly wasn't in those results. Polly suits a team that wants the lowest bill for notifications and prompts.

### Is Amazon Polly cheaper than Bland Speech?

Yes. Polly costs $4 per million characters for Standard, $16 for Neural, and $30 for Generative, against $40 for Bland Speech, and Polly's Standard free tier is 5 million characters a month. Long-Form costs $100 per million.

### How much does Bland Speech cost?

Bland Speech costs $0.04 per 1,000 characters, which is $40 per million, in the studio or the API. New accounts get 50,000 characters free, and credits do not expire. Enterprise pricing is custom, set with Bland's sales team.

### Does Amazon Polly sound as human as Bland Speech?

Design Arena's results of 4 August 2026 don't include Amazon Polly, so those results give no score that compares them. Bland Speech v3 ranked first of 13 AI voices at 1365 in those results.

### Can Amazon Polly clone a voice?

Through a Brand Voice engagement. AWS says a team interested in one contacts its AWS account manager. Bland Speech makes an instant clone from about 10 seconds of audio and a professional clone from 30 minutes or more.

### Does Amazon Polly stream audio?

Yes. Polly streams audio over HTTP, and its bidirectional StartSpeechSynthesisStream API runs over HTTP/2 on the Generative engine in listed regions. Bland Speech streams over HTTP chunks and a WebSocket.

### Does Bland Speech work on phone calls?

Yes. Bland built Bland Speech for phone calls, and it returns mu-law at 8 kHz as well as PCM from 8 to 48 kHz. Polly also returns mu-law and A-law at 8 kHz.
