# Nemotron 3.5 Transcribe

> Nemotron 3.5 Transcribe is a NVIDIA audio model in Nodaro for speech to text, from 4 credits per run.

Source: https://nodaro.ai/docs/models/audio/nemotron-3-5-transcribe

NVIDIA speech-to-text in 40 languages, detected automatically, with a low word error rate. On request it returns WORD-level timestamps with a speaker label per word, so its transcript can feed captions; or it translates the speech instead (to or from English: other speech into English, or English speech into one of 31 languages), with no word timings. Files up to 60 minutes and 200 MB. Priced per started minute of the file, whatever the options; the listed price is four minutes.

- Maker: NVIDIA
- Modes: Speech to text
- Features: Word timestamps, Speaker labels, translation
- API type: `nemotron-3-5-transcribe`

## Credits

| Tier | Credits |
| --- | --- |
| per 4 minutes of the file; charged per started minute, the whole run rounded up | 4 |

## Use Nemotron 3.5 Transcribe in Nodaro

- [Transcribe](https://nodaro.ai/docs/nodes/audio/transcribe)

## Similar models

## Frequently asked questions

### How many credits does Nemotron 3.5 Transcribe cost in Nodaro?

Nemotron 3.5 Transcribe costs 4 credits per run (per 4 minutes of the file; charged per started minute, the whole run rounded up).

### What can Nemotron 3.5 Transcribe do?

Nemotron 3.5 Transcribe supports speech to text. NVIDIA speech-to-text in 40 languages, detected automatically, with a low word error rate.

### Which Nodaro nodes can run Nemotron 3.5 Transcribe?

Choose Nemotron 3.5 Transcribe as the model on the Transcribe node.