doubao-tts

name	doubao-tts
version	2.1.0
description	Use Doubao speech synthesis (Volcengine TTS) to convert text to audio files. Triggers when user mentions "Doubao TTS", "Doubao speech synthesis", "doubao tts", "Volcengine TTS", "volcengine tts", "speech synthesis", "text to speech", "TTS", "generate audio", "read text aloud", or any scenario requiring Doubao/Volcengine speech synthesis API.

Doubao TTS Skill (V3)

Convert text to audio files using Volcengine Doubao speech synthesis V3 HTTP SSE unidirectional streaming API.

Get API Key (recommended, new console)

1. Login to Volcengine Console
2. Go to Doubao Speech -> Speech Synthesis Large Model -> App Management
3. Create app (or use existing app)
4. Get API Key from API Key Management page -> corresponds to DOUBAO_TTS_API_KEY

If service is not yet activated, first enable it on the Speech Synthesis Large Model page.

Legacy Console (AppID + Token)

Can be obtained at the bottom of legacy console app details page:

• APP ID -> DOUBAO_TTS_APPID
• Access Token -> DOUBAO_TTS_TOKEN

Environment Variables

Variable	Description	Recommended
DOUBAO_TTS_API_KEY	API Key (new console, X-Api-Key)
DOUBAO_TTS_APPID	AppID (legacy console, X-Api-App-Id)
DOUBAO_TTS_TOKEN	Access Token (legacy console, X-Api-Access-Key)
DOUBAO_TTS_RESOURCE_ID	Resource ID, leave empty for default seed-tts-2.0

Check if configured:

[ -n "$DOUBAO_TTS_API_KEY" ] && echo "API_KEY: set" || echo "API_KEY: not set"
[ -n "$DOUBAO_TTS_APPID" ] && echo "APPID: set" || echo "APPID: not set"
[ -n "$DOUBAO_TTS_TOKEN" ] && echo "TOKEN: set" || echo "TOKEN: not set"

When not configured, instruct user to set (prefer API Key): Set DOUBAO_TTS_API_KEY | Set DOUBAO_TTS_RESOURCE_ID

Legacy Console (AppID + Token): Set DOUBAO_TTS_APPID | Set DOUBAO_TTS_TOKEN

Usage

Call script: /var/minis/skills/doubao-tts/scripts/tts.py

# Basic usage
uv run --script --cache-dir /root/.cache/uv \
  /var/minis/skills/doubao-tts/scripts/tts.py \
  --text "Hello, welcome to Doubao speech synthesis" \
  --output /var/minis/workspace/output.mp3

# Specify voice and speech rate
uv run --script --cache-dir /root/.cache/uv \
  /var/minis/skills/doubao-tts/scripts/tts.py \
  --text "The weather is nice today" \
  --speaker zh_female_cancan_uranus_bigtts \
  --speech-rate 10 \
  --output /var/minis/workspace/output.mp3

# English
uv run --script --cache-dir /root/.cache/uv \
  /var/minis/skills/doubao-tts/scripts/tts.py \
  --text "Hello! Nice to meet you." \
  --speaker en_female_dacey_uranus_bigtts \
  --output /var/minis/workspace/output.mp3

API Details

• Endpoint: https://openspeech.bytedance.com/api/v3/tts/unidirectional/sse (SSE streaming)
• Authentication (choose one):
  • New console: Header X-Api-Key (API Key)
  • Legacy console: Header X-Api-App-Id + X-Api-Access-Key (AppID + Token)
• Resource ID: specifies the model version to use (see table below)
• Usage return: script includes X-Control-Require-Usage-Tokens-Return: text_words by default, returns billed character count (text_words) on completion

Resource ID	Description
seed-tts-1.0	Doubao speech synthesis model 1.0 character edition (default, compatible with all BV*_streaming voices)
seed-tts-1.0-concurr	Doubao speech synthesis model 1.0 concurrent edition
seed-tts-2.0	Doubao speech synthesis model 2.0 (only supports 2.0 voices)

Parameter Details

Parameter	Description
--text	Text to synthesize (required)
--output	Output file path (required)
--api-key	API Key (new console, takes precedence over APPID/TOKEN)
--appid	AppID (legacy console)
--token	Access Token (legacy console)
--speaker	Voice, default zh_female_shuangkuaisisi_uranus_bigtts (Shuangkuai Sisi 2.0)
--encoding	Format: mp3/pcm/ogg_opus, default mp3
--speech-rate	Speech rate [-50, 100], 0 is default, 100 is 2x speed
--loudness	Loudness [-50, 100], 0 is default
--sample-rate	Sample rate, default 24000
--emotion	Emotion (e.g. happy/sad/angry/narrator etc.)
--emotion-scale	Emotion scale [1, 5] (use with --emotion)
--resource-id	Resource ID (overrides environment variable)
--json	Output results in JSON format

Common Voice Reference

Doubao Speech Synthesis Model 2.0 (seed-tts-2.0, recommended)

speaker	Name	Scenario
zh_female_shuangkuaisisi_uranus_bigtts	Shuangkuai Sisi 2.0 Default	General
zh_female_cancan_uranus_bigtts	ZhiXing CanCan 2.0	Role-play
zh_female_tianmeixiaoyuan_uranus_bigtts	Sweet XiaoYuan 2.0	General
zh_female_vv_uranus_bigtts	Vivi 2.0	General, CN/JP/ID/MX-ES, dialects Sichuan-Shaanxi-Northeast
zh_female_xiaohe_uranus_bigtts	XiaoHe 2.0	General
zh_male_m191_uranus_bigtts	YunZhou 2.0	General
zh_male_taocheng_uranus_bigtts	XiaoTian 2.0	General
zh_female_kefunvsheng_uranus_bigtts	Warm Female Voice 2.0	Customer service
en_female_dacey_uranus_bigtts	Dacey	Multilingual (EN)
en_male_tim_uranus_bigtts	Tim	Multilingual (EN)

Doubao Speech Synthesis Model 1.0 (seed-tts-1.0, needs --resource-id change)

speaker	Name	Scenario
BV700_streaming	CanCan	General, supports 22 emotions
BV001_streaming	General Female Voice	General
BV002_streaming	General Male Voice	General
BV701_streaming	QingCang	Audio reading
BV503_streaming	Lively Female-Ariana	English

1.0 and 2.0 voices cannot be mixed, seed-tts-2.0 only supports *_uranus_bigtts voices

Common Emotion Values

pleased / sorry / happy / sad / angry / scared / surprised / hate / tear / narrator / storytelling

Complete Workflow

1. Check if environment variables are configured (prefer DOUBAO_TTS_API_KEY, then DOUBAO_TTS_APPID + DOUBAO_TTS_TOKEN)
2. Call tts.py script to generate audio to /var/minis/workspace/
3. Return as minis://workspace/xxx.mp3 link for user to click and play