# AI models: Accent Translation

- Updated on Jul 3, 2026  
- Published on Mar 18, 2026  
- 3 minute(s) read

### Introduction

The following document lists Sanas' Accent Translation (AT) model releases. Versions are listed newest-first. Each entry summarizes what changed and who it affects.

For desktop application releases that bundle these models, see [Product updates](https://help.sanas.ai/portalv3/docs/product-updates). For detailed benchmarking methodology and performance data, see [Sanas science blogs](/content/science/index.html).

* * *

#### AT5

|     |     |
| --- | --- |
| **Field** | **Detail** |
| Model version | 5.0 & 5.1 |
| Speech capability | Accent Translation |
| Target accent | US English |
| Source accent | Universal |
| Release status | GA |
| Minimum app version | 5.6.0 |

**What changed**

- Source accents are no longer limited to specific countries and geographies. They are now universally available and applicable across nearly all regions worldwide.

- Real-time translation of incoming audio accents to improve comprehension. Inbound Accent Translation can be used independently or alongside Outbound Accent Translation.

- Headset-based Model Detection for Accent Translation — Sanas now automatically applies the optimal voice-isolation mode based on the connected headset:

- Wired USB headsets → Near-field mode

- Bluetooth or other devices → General mode
- For AT5.1, the following feature has been added:

- Real-time translation of outgoing audio accents

* * *

#### AT4.8

|     |     |
| --- | --- |
| **Field** | **Detail** |
| Model version | 4.8 |
| Speech capability | Accent Translation |
| Target accent | UK English |
| Source accent | India, Filipino, Latin America, Africa, and the Middle East. |
| Release status | Early access<br>_Note: Enable the_ **_"Experimental Features"_** _option on the Portal to access Early Access models._ |
| Minimum app version | 4.3.14 |

**What changed**

- Production-ready UK English model. Addresses the primary issue of flat, inexpressive synthesis in the previous British (UK-EN) model.

- Supports Source accent (India, Filipino, Latin American, African, and Middle Eastern).

- Improved intelligibility for Latin American and African/Middle Eastern Source accent via an updated Middleweight Translator (quantized for production efficiency).

- 8kHz output optimized for lightweight contact center deployments, paired with improved VAD for fewer hallucinations and voice breaks.

* * *

#### AT4.6

|     |     |
| --- | --- |
| **Field** | **Detail** |
| Model version | 4.6 |
| Speech capability | Accent Translation |
| Target accent | US English |
| Source accent | India, Filipino, Latin America, Africa, and the Middle East. |
| Release status | Early access<br>_Note: Enable the_ **_"Experimental Features"_** _option on the Portal to access Early Access models._ |
| Minimum app version | 4.3.14 |

**What changed**

- Significant intelligibility gains across all supported Source accent (India, Filipino, Latin American, African, and Middle Eastern), with the largest improvements for Middle Eastern and African English speakers.

- Optimized Voice Activity Detection reduces hallucinations and voice breaks during pauses.

- 8kHz output support for low-bandwidth telephony and contact center environments.

* * *

#### AT4.4

|     |     |
| --- | --- |
| **Field** | **Detail** |
| Model version | 4.4 |
| Speech capability | Accent Translation |
| Target accent | UK English |
| Source accent | India, Filipino, Latin America, Africa, and the Middle East. |
| Release status | GA |
| Minimum app version | 4.2.8 |

**What changed**

- First UK English output model. Targets modern Received Pronunciation (RP) for broad intelligibility across the UK and European markets.

- Supports Indian, Filipino, Latin American, Middle Eastern, and African Source accent accents — validated through blind A/B evaluation with UK-based listeners who strongly preferred the RP output over the existing US model.

- Expanded African source accent coverage, trained on a proprietary dataset spanning 30+ subregional accents across North, East, West, and South Africa.

> _Info: Superseded by_ [_AT4.8_](https://help.sanas.ai/portalv3/docs/ai-models-accent-translation#at48) **_._** _AT 4.8 addresses expressiveness and synthesis quality issues identified in this model. New deployments should use AT 4.8._

* * *

#### AT4.3

|     |     |
| --- | --- |
| **Field** | **Detail** |
| Model version | 4.3 |
| Speech capability | Accent Translation |
| Target accent | US English |
| Source accent | Africa and the Middle East. |
| Release status | GA |
| Minimum app version | 4.2.8 |

**What changed**

- Significant intelligibility improvement for African and Middle Eastern Source accent over AT 4.1, reducing WER from 29.0% to 24.5% for this accent group.

- Finetuned on a Pan-African training dataset covering North, East, West, and South African regions, plus Caribbean accents.

- Uses an AT 4.0-style Middleweight Translator with 8kHz output and CPU utilization comparable to AT 2.8.

* * *

#### AT4.1

|     |     |
| --- | --- |
| **Field** | **Detail** |
| Model version | 4.1 |
| Speech capability | Accent Translation |
| Target accent | US English |
| Source accent | India, Filipino, Latin America, Africa, and the Middle East. |
| Release status | GA |
| Minimum Sanas App version | 4.2.8 |

**What changed**

- Major intelligibility improvement over AT 2.9 across Indian, Filipino, and Latin American English speakers, measurable in lower Word Error Rates (WER) for downstream ASR systems.

- Most natural-sounding AT model at time of release — improved intonation, pacing, and speaker similarity validated through large-scale blind A/B listening tests.

- New Lightweight mode runs with 35% fewer parameters, enabling deployment on thin clients and older hardware without sacrificing clarity.