# Cartesia Ink Review

> Cartesia Ink review: Streaming STT for voice agents with native turn detection. Verified pricing, features, and the strongest alternatives in Speech-to-text.

![Cartesia Ink logo](https://www.versusref.com/logos/cartesia-ink.png)

Streaming STT for voice agents with native turn detection

Among the 43 speech-to-text tools we track, Cartesia Ink has the 12th-widest language coverage - a fit for multilingual and localization projects.

From $5/mo · facts verified Jul 20, 2026

## What we know about Cartesia Ink

Cartesia Ink sits in the speech-to-text apis category, where it is streaming STT for voice agents with native turn detection. We keep this Cartesia Ink profile grounded in primary sources, each fact dated to when we last confirmed it.

On pricing, Cartesia Ink starts at $5 per month for its entry tier. That is the sticker rate: real production cost usually runs higher once you add a language model, a voice provider, and telephony minutes.

On capabilities, Cartesia Ink covers word-level timestamps, custom vocabulary / keyterm boosting, soc 2 type ii, self-host / on-prem option, and websocket streaming api, and does not offer entity detection, sentiment analysis, and summarization endpoint. Each of those is verified against Cartesia Ink's own docs or dashboard, not marketing copy.

For compliance, with Cartesia Ink: SOC 2 Type II is in place. If you are in a regulated space, confirm the current posture with Cartesia Ink before you commit, since these change plan by plan.

Cartesia Ink ranks the 12th-widest language coverage of the 43 speech-to-text tools we track, so where it lands for you depends on which of those matters more.

In total we track 21 verified facts for Cartesia Ink today, each linking the primary source it came from so you can check our work - and vendor claims we have not measured ourselves are labeled as such on the Cartesia Ink fact sheet below.

## Fact sheet

### Pricing

| Fact | Value | Verified | Source |
| --- | --- | --- | --- |
| Pricing model | credits | Jul 20 | [source](https://cartesia.ai/pricing) |
| Cheapest paid plan | $5 | Jul 20 | [source](https://cartesia.ai/pricing) |
| Free tier quota | 20K credits/mo (~1h 51m Ink-2 STT) | Jul 20 | [source](https://cartesia.ai/pricing) |
| Concurrency on base plan | Free: 8, Pro: 12 concurrent STT requests | Jul 20 | [source](https://cartesia.ai/pricing) |

### Capabilities

| Fact | Value | Verified | Source |
| --- | --- | --- | --- |
| Streaming latency (vendor-claimed) | ~88 ms | Jul 20 | [source](https://www.cartesia.ai/ink) |
| Languages supported | 99 languages | Jul 20 | [source](https://docs.cartesia.ai/build-with-cartesia/stt/older-models.md) |
| Speaker diarization | ✗  Not available | Jul 20 | [source](https://docs.cartesia.ai/api-reference/stt/transcribe.md) |
| PII redaction | ✗  Not available | Jul 20 | [source](https://docs.cartesia.ai/api-reference/stt/transcribe.md) |
| Word-level timestamps | ✓  Yes | Jul 20 | [source](https://docs.cartesia.ai/api-reference/stt/transcribe.md) |
| Custom vocabulary / keyterm boosting | ✓  Yes | Jul 20 | [source](https://docs.cartesia.ai/use-the-api/stt/keyterms.md) |
| Entity detection | ✗  No | Jul 20 | [source](https://docs.cartesia.ai/build-with-cartesia/stt/latest.md) |
| Sentiment analysis | ✗  No | Jul 20 | [source](https://docs.cartesia.ai/llms.txt) |
| Summarization endpoint | ✗  No | Jul 20 | [source](https://docs.cartesia.ai/llms.txt) |
| Speech translation | ✗  No | Jul 20 | [source](https://docs.cartesia.ai/llms.txt) |

### Compliance & trust

| Fact | Value | Verified | Source |
| --- | --- | --- | --- |
| HIPAA BAA available | ✓  Yes | Jul 20 | [source](https://www.cartesia.ai/industries/healthcare) |
| SOC 2 Type II | ✓  Yes | Jul 20 | [source](https://www.cartesia.ai/industries/healthcare) |
| Self-host / on-prem option | ✓  Yes | Jul 20 | [source](https://docs.cartesia.ai/self-hosted/introduction) |

### Build experience

| Fact | Value | Verified | Source |
| --- | --- | --- | --- |
| Official SDKs | Python, JS/TS | Jul 20 | [source](https://docs.cartesia.ai/tools/client-libraries.md) |
| Max file size / duration | Batch: any length (server chunks long files) | Jul 20 | [source](https://docs.cartesia.ai/api-reference/stt/transcribe.md) |
| Supported audio formats | Batch: flac,m4a,mp3,mp4,ogg,wav,webm; stream: raw PCM | Jul 20 | [source](https://docs.cartesia.ai/api-reference/stt/transcribe.md) |
| Websocket streaming API | ✓  Yes | Jul 20 | [source](https://docs.cartesia.ai/api-reference/stt/turns/websocket.md) |

## Cartesia Ink head-to-head

| Comparison | Record | Per use case |
| --- | --- | --- |
| [Cartesia Ink vs Deepgram](https://www.versusref.com/stt/cartesia-ink-vs-deepgram/) | won 1 · lost 4 · tied 2 | Call Centers: lost; Developers: lost; Dictation: tie; Medical: lost; Meetings: lost; Self-Hosted: tie; Voice Agents: won |

Source: https://www.versusref.com/stt/tools/cartesia-ink/