> ## Documentation Index
> Fetch the complete documentation index at: https://docs.langcraft.world/llms.txt
> Use this file to discover all available pages before exploring further.

# Early-childhood speech analysis

> Pronunciation analysis designed for young learners

Use the Nova 1 model for early-childhood speech analysis. It is designed for
young speakers from approximately six months to eight years old and accounts
for variation in children’s voices, pronunciations, and developing speech
patterns.

## Select Nova 1

Set the `model` field in your request:

```text theme={null}
model=nova-1
```

For example:

```bash theme={null}
curl -s -X POST https://api.langcraft.world/v1/speech/analyze \
  -H "x-api-key: $LANGCRAFT_API_KEY" \
  -F "audio=@utterance.wav" \
  -F "reference_text=the red house" \
  -F "lang=en" \
  -F "model=nova-1"
```

All other request fields and the response format remain unchanged.

Alternatively, sending `speaker_profile=early_childhood` also selects Nova 1:

```text theme={null}
speaker_profile=early_childhood
```

Send either selector. If both fields are included, `model` must be `nova-1`.
Nova 1 requires supplied reference text or reference phones.

## Supported languages

Nova 1 currently supports:

* English (`en`)
* French (`fr`)
* German (`de`)
* Portuguese (`pt`)
* Spanish (`es`)

We plan to expand Nova 1 to additional languages.