From 6d269edc739f71ee6f025210191939bded477d01 Mon Sep 17 00:00:00 2001
From: "mintlify[bot]" <109931778+mintlify[bot]@users.noreply.github.com>
Date: Tue, 26 May 2026 15:10:32 +0000
Subject: [PATCH] docs: add Cartesia Sonic 3.5 voice model
---
releases/notes/26.05.mdx | 16 +++++++++++++++-
tools/classes.mdx | 2 +-
tools/classes/voice.mdx | 9 ++++++---
voice/add-a-new-voice.mdx | 2 +-
4 files changed, 23 insertions(+), 6 deletions(-)
diff --git a/releases/notes/26.05.mdx b/releases/notes/26.05.mdx
index 1d5dfce1..6f1ef7dd 100644
--- a/releases/notes/26.05.mdx
+++ b/releases/notes/26.05.mdx
@@ -3,7 +3,7 @@ title: 05.2026
description: May 2026 release notes.
---
-The **May 2026** PolyAI Agent Studio release ships a redesigned **Conversations** workspace built for high-volume review and adds a one-click **Tripleseat** integration for restaurant agents.
+The **May 2026** PolyAI Agent Studio release ships a redesigned **Conversations** workspace built for high-volume review, adds a one-click **Tripleseat** integration for restaurant agents, and brings Cartesia's **Sonic 3.5** voice model to Agent Studio.
Expand the items below for details:
@@ -23,6 +23,20 @@ See [Conversation review](/analytics/conversations/review), [Views](/analytics/c
+
+
+Cartesia's latest TTS model, **Sonic 3.5**, is now selectable for any Cartesia voice. Pick it from the model dropdown when adding or editing a Cartesia voice, or set `model_id="sonic-3.5"` in [`CartesiaVoice`](/tools/classes/voice#example-cartesia).
+
+Sonic 3.5 inherits all Sonic 3 behavior:
+
+- 10-emotion set via the `emotion` parameter
+- `volume` control (e.g. `0.5`–`2.0`)
+- Speed range `0.6`–`1.5`
+
+No code changes are required for existing Cartesia voices — switch the model ID to opt in.
+
+
+
Connect PolyAI to **Tripleseat** to capture event and large-party leads from voice, webchat, and SMS conversations — without writing custom code. Projects on the PLG restaurant template can connect Tripleseat through one-click OAuth from the Integrations page; other projects continue to be supported through Managed Services.
diff --git a/tools/classes.mdx b/tools/classes.mdx
index a9f8d927..d5039ce2 100644
--- a/tools/classes.mdx
+++ b/tools/classes.mdx
@@ -85,7 +85,7 @@ VoiceWeighting(
| `provider_voice_id` | The Cartesia voice to use |
| `speed` | Speech rate: -1.0 (slowest) to 1.0 (fastest) |
| `emotions` | List of `Emotion` objects (see [voice reference](/tools/classes/voice)) |
- | `model_id` | `"sonic"` or `"sonic-preview"` |
+ | `model_id` | `"sonic"`, `"sonic-preview"`, `"sonic-3"`, or `"sonic-3.5"` (latest) |
Configures voice settings for [PlayHT](https://docs.play.ht/reference/api-getting-started) TTS.
diff --git a/tools/classes/voice.mdx b/tools/classes/voice.mdx
index 08ab9840..4dcdc54e 100644
--- a/tools/classes/voice.mdx
+++ b/tools/classes/voice.mdx
@@ -48,7 +48,7 @@ conv.set_voice(
emotions=[
Emotion(EmotionKind.POSITIVITY, EmotionIntensity.HIGH)
],
- model_id="sonic" # or any Cartesia-compatible identifier e.g. "sonic-3-2025-10-27"
+ model_id="sonic" # also: "sonic-preview", "sonic-3", "sonic-3.5", or a dated identifier e.g. "sonic-3-2025-10-27"
)
)
```
@@ -59,10 +59,13 @@ conv.set_voice(
- `EmotionKind`: `ANGER`, `POSITIVITY`, `SURPRISE`
- `EmotionIntensity`: `LOWEST`, `LOW`, `HIGH`, `HIGHEST`
-**Sonic 3 parameters**: When using a Sonic 3 model ID, the following additional parameters are supported:
+**Sonic 3 parameters**: When using a Sonic 3 model ID (`sonic-3` or `sonic-3.5`), the following additional parameters are supported:
- `volume` (float, optional) – controls output volume (e.g. 0.5–2.0).
-- `emotion` (str, optional) – emotion string (e.g. `"happy"`).
+- `emotion` (str, optional) – emotion string (e.g. `"happy"`). Sonic 3 models support a 10-emotion set.
- `language` (str, optional) – language code (e.g. `"en"`).
+- `speed` – on Sonic 3 models, the effective speed range is `0.6`–`1.5`.
+
+`sonic-3.5` is the latest Cartesia model and inherits all Sonic 3 behavior. Use it for the most natural, expressive output; fall back to `sonic-3` or `sonic` if you need an earlier model for parity with existing voices.
### Example: Rime
diff --git a/voice/add-a-new-voice.mdx b/voice/add-a-new-voice.mdx
index 2a3b1c0f..3334bdc2 100644
--- a/voice/add-a-new-voice.mdx
+++ b/voice/add-a-new-voice.mdx
@@ -30,7 +30,7 @@ conv.set_voice(
emotions=[
Emotion(EmotionKind.POSITIVITY, EmotionIntensity.HIGH)
],
- model_id="sonic" # or "sonic-preview"
+ model_id="sonic-3.5" # also: "sonic", "sonic-preview", "sonic-3"
)
)
```