diff --git a/chapters/audio-intelligence/index.mdx b/chapters/audio-intelligence/index.mdx
index 7e6568a..f702893 100644
--- a/chapters/audio-intelligence/index.mdx
+++ b/chapters/audio-intelligence/index.mdx
@@ -40,6 +40,14 @@ Use these capabilities alongside Live or Pre-recorded STT to automate workflows
Automatically redact names, emails, vehicle IDs, and other PII in pre-recorded transcripts.
+
+ Group words into sentences with timing for more readable pre-recorded transcripts.
+
+
+
+In addition to getting the transcription split by utterances, you can request semantic sentence segmentation for more readable results. This feature is only available for **pre-recorded** transcription.
-You can get translated sentences by enabling both `sentences` and `translation`! You'll receive sentences output for the original transcript, and each `translation` result will also contain sentences in the translated language.
+You can get translated sentences by enabling both `sentences` and [translation](/chapters/audio-intelligence/translation)! You'll receive sentences output for the original transcript, and each `translation` result will also contain sentences in the translated language.
-```json request data
+```json Pre-recorded
{
+ "audio_url": "YOUR_AUDIO_URL",
"sentences": true
}
```
@@ -36,4 +41,3 @@ The result will contain a `sentences` key (in addition to `utterances`):
}
```
-
diff --git a/chapters/audio-intelligence/translation.mdx b/chapters/audio-intelligence/translation.mdx
index ca66532..8fec2f9 100644
--- a/chapters/audio-intelligence/translation.mdx
+++ b/chapters/audio-intelligence/translation.mdx
@@ -9,7 +9,7 @@ import LiveBadge from "/snippets/badges/live.mdx"
-The **Translation** model generates translations of your transcriptions to one or more targeted languages. If [subtitles](/chapters/audio-intelligence/subtitles) and/or [sentences](/chapters/pre-recorded-stt/features/sentences) are enabled, the translations will also include translated results for them.
+The **Translation** model generates translations of your transcriptions to one or more targeted languages. If [subtitles](/chapters/audio-intelligence/subtitles) and/or [sentences](/chapters/audio-intelligence/sentences) are enabled, the translations will also include translated results for them.
You can translate your transcription to **multiple languages** in a single API call.
The list of the languages covered by the Translation feature are listed in [Supported Languages](/chapters/language/supported-languages).
@@ -204,8 +204,7 @@ The transcription result will contain a `"translation"` key with the output of t
- If you enabled the `subtitles` generation, those will also benefits from the
- translation model.
+ If you enabled [subtitles](/chapters/audio-intelligence/subtitles) or [sentences](/chapters/audio-intelligence/sentences), those will also benefit from the translation model.
## Best practices
diff --git a/chapters/pre-recorded-stt/benchmarking.mdx b/chapters/how-to-use-gladia/benchmarking.mdx
similarity index 98%
rename from chapters/pre-recorded-stt/benchmarking.mdx
rename to chapters/how-to-use-gladia/benchmarking.mdx
index 1b57237..36f9eff 100644
--- a/chapters/pre-recorded-stt/benchmarking.mdx
+++ b/chapters/how-to-use-gladia/benchmarking.mdx
@@ -149,7 +149,7 @@ Typical failure cases:
public datasets that match your use case.
-For a broader methodology view, see [this benchmark guide](/chapters/pre-recorded-stt/benchmarking), especially the evaluation-goal section above when mapping use cases to dataset types.
+For a broader methodology view, see [this benchmark guide](/chapters/how-to-use-gladia/benchmarking), especially the evaluation-goal section above when mapping use cases to dataset types.
## 4. Interpret results carefully
diff --git a/chapters/introduction/playground.mdx b/chapters/how-to-use-gladia/playground.mdx
similarity index 100%
rename from chapters/introduction/playground.mdx
rename to chapters/how-to-use-gladia/playground.mdx
diff --git a/chapters/introduction/index.mdx b/chapters/introduction/index.mdx
index e73570f..2ab9614 100644
--- a/chapters/introduction/index.mdx
+++ b/chapters/introduction/index.mdx
@@ -43,7 +43,7 @@ in both Real-time and asynchronous ways, with audio intelligence tools to extrac
Test gladia quickly on our playground.
diff --git a/chapters/introduction/models.mdx b/chapters/introduction/models.mdx
new file mode 100644
index 0000000..d5c7c64
--- /dev/null
+++ b/chapters/introduction/models.mdx
@@ -0,0 +1,181 @@
+---
+title: Models
+description: Compare Gladia speech-to-text models and choose the right one for your use case.
+---
+
+Pick a model with `model` in your request. Omit it to use the default: **`solaria-1`**.
+
+## Choose your model
+
+
+
+ Highest accuracy on European real-world audio. Pre-recorded only.
+
+
+ Default generalist. 100+ languages, async and live, code switching.
+
+
+
+## Comparison
+
+| | **Solaria-3** | **Solaria-1** (default) |
+| --- | --- | --- |
+| **Best for** | European real-world / business audio | Global coverage across any domain |
+| **Modes** | Pre-recorded only | Pre-recorded + live |
+| **Languages** | EN, FR, DE, ES, IT | 100+ languages |
+| **Code switching** | No | Yes |
+| **Benchmarks** | Optimized for real-world European production audio. [See WER results →](https://www.gladia.io/competitors/benchmarks) | Optimized for clean speech, live streaming, and multilingual coverage. [See WER results →](https://www.gladia.io/competitors/benchmarks) |
+| **Audio Intelligence** | All add-ons | All add-ons |
+| **Pricing** | [Same plan rates](https://www.gladia.io/pricing) | [Same plan rates](https://www.gladia.io/pricing) |
+
+## Select a model in your API call
+
+Set `model` to `"solaria-3"` or `"solaria-1"`:
+
+
+
+```javascript JavaScript
+import { GladiaClient } from "@gladiaio/sdk";
+
+const gladiaClient = new GladiaClient({ apiKey: "YOUR_GLADIA_API_KEY" });
+
+const transcription = await gladiaClient.preRecorded().transcribe(
+ "YOUR_AUDIO_URL_OR_LOCAL_PATH",
+ {
+ model: "solaria-3",
+ language_config: {
+ languages: ["fr"],
+ },
+ }
+);
+```
+
+```python Python
+from gladiaio_sdk import GladiaClient
+
+gladia_client = GladiaClient(api_key="YOUR_GLADIA_API_KEY").prerecorded()
+
+transcription = gladia_client.transcribe(
+ "YOUR_AUDIO_URL_OR_LOCAL_PATH",
+ {
+ "model": "solaria-3",
+ "language_config": {
+ "languages": ["fr"],
+ },
+ },
+)
+```
+
+```bash cURL
+curl --request POST 'https://api.gladia.io/v2/pre-recorded' \
+ --header 'x-gladia-key: YOUR_GLADIA_API_KEY' \
+ --header 'Content-Type: application/json' \
+ --data-raw '{
+ "audio_url": "YOUR_AUDIO_URL",
+ "model": "solaria-3",
+ "language_config": {
+ "languages": ["fr"]
+ }
+ }'
+```
+
+
+
+
+ With `"solaria-3"`, there is no code switching. You can pass languages in `language_config.languages` (for example `["fr"]`, or several from EN/FR/DE/ES/IT) so the model can pick among those covered languages. Once the language is detected, it sticks to it for the rest of the file.
+
+
+## Solaria-3
+
+Highest accuracy on noisy, conversational European audio (calls, meetings, field recordings).
+
+### When to use
+
+Use **`solaria-3`** if:
+
+- Audio is **pre-recorded**
+- Language is **EN, FR, DE, ES, or IT**
+- You prioritize **accuracy on real production audio**
+- You can constrain detection to the **covered languages** (no mid-file switching)
+
+Use Solaria-1 instead for live streaming, code switching, languages outside those five, or clean / formal speech.
+
+### Get started
+
+
+
+ Transcribe files with `model: "solaria-3"`.
+
+
+ Call Solaria-3 from JavaScript or Python.
+
+
+
+## Solaria-1
+
+Maximum language coverage. Works async and live, with code switching.
+
+### When to use
+
+Use **`solaria-1`** (or omit `model`) if:
+
+- You need **live / real-time** transcription
+- You need **100+ languages** or **code switching**
+- Audio is **clean read-speech** or formal / institutional
+
+Ideal default for voice agents, live captions, and global multilingual products.
+
+### Get started
+
+
+
+ Stream audio in real time.
+
+
+ Batch-transcribe with the default model.
+
+
+ Integrate with the official SDKs.
+
+
+
+
+
+ Handle mixed-language conversations.
+
+
+ Full language list.
+
+
+
+
+ Compare Solaria-3 and Solaria-1 WER across real customer audio and public datasets on our [open benchmarks page](https://www.gladia.io/competitors/benchmarks). To measure accuracy on your own traffic, see [Benchmarking](/chapters/how-to-use-gladia/benchmarking).
+
diff --git a/chapters/pre-recorded-stt/audio-intelligence.mdx b/chapters/pre-recorded-stt/audio-intelligence.mdx
index aab4645..986c681 100644
--- a/chapters/pre-recorded-stt/audio-intelligence.mdx
+++ b/chapters/pre-recorded-stt/audio-intelligence.mdx
@@ -34,6 +34,13 @@ Audio intelligence turns raw speech into structured, useful data on top of trans
>
Generate ready-to-use subtitle files in SRT or VTT formats.
+
+ Group words into sentences with timing for more readable pre-recorded transcripts.
+
Group words into sentences with timing for better readability and parsing.
diff --git a/chapters/pre-recorded-stt/recommended-parameters.mdx b/chapters/pre-recorded-stt/recommended-parameters.mdx
index fccc73c..3ce2002 100644
--- a/chapters/pre-recorded-stt/recommended-parameters.mdx
+++ b/chapters/pre-recorded-stt/recommended-parameters.mdx
@@ -48,7 +48,7 @@ For apps that record and process meetings — team stand-ups, board sessions, 1-
| `diarization_config.min_speakers` / `max_speakers` | Set a range (e.g. `2`-`10`) | Meeting size varies — a range lets the model adapt without over- or under-splitting speakers. |
| `summarization` | `true` | Generates a summary for quick review. Use `bullet_points` type for action-item style output. See [Summarization](/chapters/audio-intelligence/summarization). |
| `named_entity_recognition` | `true` | Surfaces people, organizations, dates, and other key entities mentioned during the meeting. See [NER](/chapters/audio-intelligence/named-entity-recognition). |
-| `sentences` | `true` | Produces well-segmented, readable output suitable for meeting minutes. See [Sentences](/chapters/pre-recorded-stt/features/sentences). |
+| `sentences` | `true` | Produces well-segmented, readable output suitable for meeting minutes. See [Sentences](/chapters/audio-intelligence/sentences). |
| `language_config.languages` | Set explicitly | Meeting language is almost always known in advance — setting it avoids detection overhead. |
| `custom_vocabulary` | `true` | Add company-specific terms, project names, and participant names for better accuracy. |
@@ -80,7 +80,7 @@ For long-form audio with multiple speakers the focus is on **readability** and *
| -------------------------------------------------- | -------------------------- | --------------------------------------------------------------------------------------------------------------------------------- |
| `diarization` | `true` | Essential for multi-speaker content. |
| `diarization_config.min_speakers` / `max_speakers` | Set a range (e.g. `2`-`4`) | Provides a flexible hint when the exact count varies across episodes. |
-| `sentences` | `true` | Produces well-segmented, readable output suitable for publishing. See [Sentences](/chapters/pre-recorded-stt/features/sentences). |
+| `sentences` | `true` | Produces well-segmented, readable output suitable for publishing. See [Sentences](/chapters/audio-intelligence/sentences). |
| `custom_vocabulary` | `true` | Add recurring guest names, show-specific terms, and brand names. |
| `language_config.languages` | Set explicitly | Podcast language is almost always known in advance. |
diff --git a/docs.json b/docs.json
index c89e4ab..a13b564 100644
--- a/docs.json
+++ b/docs.json
@@ -29,8 +29,14 @@
"pages": [
"chapters/introduction/index",
"chapters/introduction/getting-started",
- "chapters/introduction/playground",
- "chapters/pre-recorded-stt/benchmarking"
+ "chapters/introduction/models"
+ ]
+ },
+ {
+ "group": "How to use Gladia",
+ "pages": [
+ "chapters/how-to-use-gladia/playground",
+ "chapters/how-to-use-gladia/benchmarking"
]
},
{
@@ -42,7 +48,6 @@
"pages": [
"chapters/pre-recorded-stt/quickstart",
"chapters/pre-recorded-stt/audio-intelligence",
- "chapters/pre-recorded-stt/features/sentences",
"chapters/pre-recorded-stt/recommended-parameters"
]
},
@@ -74,6 +79,7 @@
"chapters/audio-intelligence/speaker-diarization",
"chapters/audio-intelligence/translation",
"chapters/audio-intelligence/subtitles",
+ "chapters/audio-intelligence/sentences",
"chapters/audio-intelligence/custom-vocabulary",
"chapters/audio-intelligence/custom-spelling",
"chapters/audio-intelligence/named-entity-recognition",
@@ -306,6 +312,18 @@
}
},
"redirects": [
+ {
+ "source": "/chapters/introduction/playground",
+ "destination": "/chapters/how-to-use-gladia/playground"
+ },
+ {
+ "source": "/chapters/pre-recorded-stt/benchmarking",
+ "destination": "/chapters/how-to-use-gladia/benchmarking"
+ },
+ {
+ "source": "/chapters/pre-recorded-stt/features/sentences",
+ "destination": "/chapters/audio-intelligence/sentences"
+ },
{
"source": "/api-reference/transcription-flow",
"destination": "/api-reference/pre-recorded-flow"