惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

MyScale Blog
MyScale Blog
博客园 - 司徒正美
A
About on SuperTechFans
Vercel News
Vercel News
H
Hackread – Cybersecurity News, Data Breaches, AI and More
爱范儿
爱范儿
I
InfoQ
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
博客园_首页
Google DeepMind News
Google DeepMind News
T
Tailwind CSS Blog
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
F
Fortinet All Blogs
S
SegmentFault 最新的问题
阮一峰的网络日志
阮一峰的网络日志
D
Docker
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
G
Google Developers Blog
Stack Overflow Blog
Stack Overflow Blog
M
MIT News - Artificial intelligence
Jina AI
Jina AI
H
Help Net Security
量子位
IT之家
IT之家

Recent Commits to openclaw:main

test: merge chat side-result checks · openclaw/openclaw@ddd2c2a test: merge cron history checks · openclaw/openclaw@f7eb746 test: merge responsive navigation shell checks · openclaw/openclaw@c2e4b47 docs(changelog): add codex oauth fixes · openclaw/openclaw@628e6cd test: merge navigation routing cases · openclaw/openclaw@5d8cecb Tests: mock channel registry bundled fallback · openclaw/openclaw@2b08233 Secrets: avoid broad web search discovery for single plugin config · openclaw/openclaw@a464f59 test: merge config view browser checks · openclaw/openclaw@20cf511 fix(status): align oauth health with runtime · openclaw/openclaw@eed7116 feat: add macOS screen snapshots for monitor preview (#67954) thanks … · openclaw/openclaw@f377db1 fix: report shared auth scopes in hello-ok (#67810) thanks @BunsDev · openclaw/openclaw@0b6c39b Auto-reply: avoid eager bundled route fallback · openclaw/openclaw@3ea1bf4 Tests: narrow session binding contract setup · openclaw/openclaw@54e4e16 fix(macOS): enable undo/redo in webchat composer text input (#34962) · openclaw/openclaw@00951dc Tests: speed up channel setup promotion · openclaw/openclaw@82b529a Docs: refresh agent instructions · openclaw/openclaw@5775fe2 fix(auth): serialize OAuth refresh across agents to fix #26322 (#67876) · openclaw/openclaw@8e79080 test: allow ollama public surface boundary test · openclaw/openclaw@7d4f1a6 Docs: add test performance guardrails · openclaw/openclaw@89706d3 Tests: restore context-engine usage proof · openclaw/openclaw@e4c4f95 Tests: slim context engine runtime coverage · openclaw/openclaw@74c198f ci: retry failed custom checkouts · openclaw/openclaw@0ee5baf test: trim duplicate provider auth onboarding cases · openclaw/openclaw@1ffc02e matrix: fix sessions_spawn --thread subagent session spawning (#67643) · openclaw/openclaw@1ce2596 test: reduce auth choice fixture churn · openclaw/openclaw@857b9cd test: mock health status config boundaries · openclaw/openclaw@9d5ab4a test: mock onboard config io boundary · openclaw/openclaw@299694d test: mock legacy state plugin boundaries · openclaw/openclaw@2713089 test: mock channel install boundaries · openclaw/openclaw@b945248 test: mock doctor preview channel boundaries · openclaw/openclaw@b1a3ad4
docs(providers): rewrite Gradium TTS reference with verif...
vincentkoc · 2026-05-10 · via Recent Commits to openclaw:main

@@ -2,21 +2,50 @@

22

summary: "Use Gradium text-to-speech in OpenClaw"

33

read_when:

44

- You want Gradium for text-to-speech

5-

- You need Gradium API key or voice configuration

5+

- You need Gradium API key, voice, or directive token configuration

66

title: "Gradium"

77

---

889-

Gradium is a bundled text-to-speech provider for OpenClaw. It can generate normal audio replies, voice-note-compatible Opus output, and 8 kHz u-law audio for telephony surfaces.

9+

[Gradium](https://gradium.ai) is a bundled text-to-speech provider for OpenClaw. The plugin can render normal audio replies (WAV), voice-note-compatible Opus output, and 8 kHz u-law audio for telephony surfaces.

10+11+

| Property | Value |

12+

| ------------- | ------------------------------------ |

13+

| Provider id | `gradium` |

14+

| Auth | `GRADIUM_API_KEY` or config `apiKey` |

15+

| Base URL | `https://api.gradium.ai` (default) |

16+

| Default voice | `Emma` (`YTpq7expH9539ERJ`) |

10171118

## Setup

121913-

Create a Gradium API key, then expose it to OpenClaw:

20+

Create a Gradium API key, then expose it to OpenClaw with either an env var or the config key.

142115-

```bash

16-

export GRADIUM_API_KEY="gsk_..."

17-

```

22+

<Tabs>

23+

<Tab title="Env var">

24+

```bash

25+

export GRADIUM_API_KEY="gsk_..."

26+

```

27+

</Tab>

28+29+

<Tab title="Config key">

30+

```json5

31+

{

32+

messages: {

33+

tts: {

34+

auto: "always",

35+

provider: "gradium",

36+

providers: {

37+

gradium: {

38+

apiKey: "${GRADIUM_API_KEY}",

39+

},

40+

},

41+

},

42+

},

43+

}

44+

```

45+

</Tab>

46+

</Tabs>

184719-

You can also store the key in config under `messages.tts.providers.gradium.apiKey`.

48+

The plugin checks the resolved `apiKey` first and falls back to the `GRADIUM_API_KEY` environment variable.

20492150

## Config

2251

@@ -38,6 +67,14 @@ You can also store the key in config under `messages.tts.providers.gradium.apiKe

3867

}

3968

```

406970+

| Key | Type | Description |

71+

| ---------------------------------------- | ------ | --------------------------------------------------------------------------------------------- |

72+

| `messages.tts.providers.gradium.apiKey` | string | Resolved API key. Supports `${ENV}` and secret refs. |

73+

| `messages.tts.providers.gradium.baseUrl` | string | Override the API origin. Trailing slashes are stripped. Defaults to `https://api.gradium.ai`. |

74+

| `messages.tts.providers.gradium.voiceId` | string | Default voice id used when no directive override is present. |

75+76+

The output audio format is selected automatically by the runtime based on the target surface and is not configurable from `openclaw.json`. See [Output](#output) below.

77+4178

## Voices

42794380

| Name | Voice ID |

@@ -52,11 +89,33 @@ You can also store the key in config under `messages.tts.providers.gradium.apiKe

52895390

Default voice: Emma.

549192+

### Per-message voice override

93+94+

When the active speech policy allows voice overrides, you can switch voices inline using a directive token. All of these resolve to the same `voiceId` override:

95+96+

```text

97+

/voice:LFZvm12tW_z0xfGo

98+

/voice_id:LFZvm12tW_z0xfGo

99+

/voiceid:LFZvm12tW_z0xfGo

100+

/gradium_voice:LFZvm12tW_z0xfGo

101+

/gradiumvoice:LFZvm12tW_z0xfGo

102+

```

103+104+

If the speech policy disables voice overrides, the directive is consumed but ignored.

105+55106

## Output

5610757-

- Audio-file replies use WAV.

58-

- Voice-note replies use Opus and are marked voice-compatible.

59-

- Telephony synthesis uses `ulaw_8000` at 8 kHz.

108+

The runtime picks the output format from the target surface. The provider does not synthesize other formats today.

109+110+

| Target | Format | File ext | Sample rate | Voice-compatible flag |

111+

| -------------- | ----------- | -------- | ----------- | --------------------- |

112+

| Standard audio | `wav` | `.wav` | provider | no |

113+

| Voice note | `opus` | `.opus` | provider | yes |

114+

| Telephony | `ulaw_8000` | n/a | 8 kHz | n/a |

115+116+

## Auto-select order

117+118+

Among configured TTS providers, Gradium's auto-select order is `30`. See [Text-to-Speech](/tools/tts) for how OpenClaw picks the active provider when `messages.tts.provider` is not pinned.

6011961120

## Related

62121