> ## Documentation Index
> Fetch the complete documentation index at: https://superwhisper-docs-cloud-model-rate-limits.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Voice

> Compare Superwhisper's voice-to-text models — cloud and local options for fast, accurate transcription in 100+ languages.

Voice models take your recorded voice and transcribe them into text. There are two types of voice models: local and cloud-based. Local (or Offline) models run on your machine. Cloud models are hosted either by Superwhisper or another provider.

## Superwhisper (Cloud)

These are the models that are hosted in the Cloud by Superwhisper. They are optimized for low latency and accuracy, and are available for use in 100+ languages.

| Name             | Lang  | Translate | Speed | Accuracy | Size  | License |
| ---------------- | ----- | --------- | ----- | -------- | ----- | ------- |
| S1-Voice (Cloud) | multi | ✔         | 10    | 9        | Cloud | Pro     |
| Ultra (Cloud)    | multi | ✔         | 9     | 9        | Cloud | Pro     |

## Whisper Models (Local)

These models run on your machine, allowing you to transcribe audio without relying on external services.
They are based on the [Whisper](https://github.com/openai/whisper) model series from OpenAI. They run locally using [whisper.cpp](https://github.com/ggerganov/whisper.cpp).

| Name                     | Lang | Translate | Speed | Accuracy | Size   | License |
| ------------------------ | ---- | --------- | ----- | -------- | ------ | ------- |
| Ultra V3 Turbo           | all  | ✔         | 8     | 8        | 1.6 GB | Pro     |
| Ultra                    | all  | ✔         | 6     | 10       | 3 GB   | Pro     |
| Ultra V3 Turbo (Chinese) | zh   | ✖         | 8     | 8        | 1.6 GB | Pro     |
| Pro                      | all  | ✔         | 7     | 8        | 1.5 GB | Pro     |
| Pro (English)            | en   | ✖         | 7     | 8        | 1.5 GB | Pro     |
| Standard                 | all  | ✔         | 8     | 5        | 500 MB | Free    |
| Standard (English)       | en   | ✖         | 8     | 5        | 500 MB | Free    |
| Nano                     | all  | ✔         | 9     | 3        | 150 MB | Free    |
| Nano (English)           | en   | ✖         | 9     | 3        | 150 MB | Free    |
| Fast                     | all  | ✔         | 10    | 1        | 75 MB  | Free    |
| Fast (English)           | en   | ✖         | 10    | 1        | 75 MB  | Free    |

## Nvidia Parakeet (Local)

They are based on the [Parakeet](https://huggingface.co/nvidia/parakeet-tdt-0.6b-v2) model. They run locally (on your laptop) using Argmax's [WhisperKit SDK](https://github.com/argmaxinc/whisperkit). They are extremely fast and run in parallel over long recordings. Drawbacks here are they do tend to struggle with punctuation and have minor hallucination issues with single word recordings.

| Name                   | Lang  | Translate | Speed | Accuracy | Size   | License |
| ---------------------- | ----- | --------- | ----- | -------- | ------ | ------- |
| Parakeet               | en    | ✖         | 10    | 8        | 476 MB | Pro     |
| Parakeet Multilanguage | multi | ✖         | 10    | 8        | 494 MB | Pro     |

## Deepgram (Cloud)

The Nova series of models are a Cloud model hosted by Deepgram.

| Name         | Lang  | Trans. | Speed | Accuracy | Size  | License |
| ------------ | ----- | ------ | ----- | -------- | ----- | ------- |
| Nova 3       | multi | –      | 7     | 8        | Cloud | Pro     |
| Nova 2       | multi | –      | 7     | 7        | Cloud | Pro     |
| Nova Medical | en    | –      | 10    | 7        | Cloud | Pro     |

## Fair Use & Rate Limits

Superwhisper's hosted cloud voice models share the same fair-use rate limits as the hosted cloud language models. These are abuse-prevention limits applied across all hosted cloud models combined, so normal interactive dictation will not reach them.

<Note>
  See [Fair Use & Rate Limits](../models/language#fair-use-rate-limits) on the Language Models page for the current thresholds and details.
</Note>