How to Use Qwen3-TTS 1.7B in Claude Code
Qwen3-TTS 1.7B turns scripts into multilingual narration with instructions for tone, pace, and emotion. Unlike Qwen3-TTS 0.6B CustomVoice, it supports instruction-based control of delivery.
Prompts to try with Qwen3-TTS 1.7B in Claude Code
Copy an example into Claude Code, or ask in your own words.
Read the project's onboarding documentation and write a 100-word introduction for new users. Use Qwen3-TTS 1.7B to narrate it in English with a patient tone and measured pace.
Find the password-reset instructions in the project and turn them into a short spoken explanation. Use Qwen3-TTS 1.7B with reassuring English delivery.
Use the project's release notes to draft a 90-word spoken update. Generate the narration with Qwen3-TTS 1.7B in English, sounding pleased but restrained.
Take the first tutorial script in the project and generate two readings with Qwen3-TTS 1.7B: one slow and instructional, one conversational. Keep the words identical.
Find the reviewed Japanese onboarding copy in the project and generate a short narration with Qwen3-TTS 1.7B. Use clear, measured delivery and retain the approved wording.
How to set up Qwen3-TTS 1.7B in Claude Code
Pick the way that suits you. Each takes about two minutes.
Use Spiral's MCP server
- Add the hosted server
This adds Spiral to the current project. Add --scope user to use it in every project.
claude mcp add --transport http spiral https://api.spiral.ai/mcp
- Sign in
Start claude, type /mcp, choose spiral and Authenticate. Sign in to Spiral in the browser and click Authorize. No API key.
Use the Spiral skill
- Send Claude Code this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run Qwen3-TTS 1.7B
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "fal", "model": "fal-ai/qwen-3-tts/text-to-speech/1.7b", "input": { "text": "I am solving the equation: x = [-b ± √(b²-4ac)] / 2a? Nobody can — it'\''s a disaster (◍•͈⌔•͈◍), very sad!" } }'
Then paste one of the prompts above: Claude Code finds Qwen3-TTS 1.7B on Spiral, shows the price and runs it.
Qwen3-TTS 1.7B providers and prices
Spiral runs Qwen3-TTS 1.7B through fal. Open a card for today's price, with every fee included.
$0.117 per 1,000 characters of text. Rounded up to the next 1,000 characters, minimum 1,000.
Settings you can ask for
Qwen3-TTS 1.7B works from your prompt. Ask Claude Code for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Voice | Vivian, Serena, Uncle_Fu, Dylan, Eric, Ryan, Aiden, Ono_Anna and 1 more | The voice to be used for speech synthesis, will be ignored if a speaker embedding is provided. Check out the [documentation](https://github.com/QwenLM/Qwen3-TTS/tree/main?tab=readme-ov-file#custom-voice-generate) for each voice's details and which language they primarily support. | |
| Language | Auto, English, Chinese, Spanish, French, German, Italian, Japanese and 3 more | Auto | The language of the voice. |
| Top k | a whole number from 0 | 50 | Top-k sampling parameter. |
| Top p | 0 to 1 | 1 | Top-p sampling parameter. |
| Temperature | a number up to 1 | 0.9 | Sampling temperature; higher => more random. |
| Repetition penalty | a number from 0 | 1.05 | Penalty to reduce repeated tokens/codes. |
| Subtalker dosample | on or off | on | Sampling switch for the sub-talker. |
| Subtalker top k | a whole number from 0 | 50 | Top-k for sub-talker sampling. |
| Subtalker top p | 0 to 1 | 1 | Top-p for sub-talker sampling. |
| Subtalker temperature | 0 to 1 | 0.9 | Temperature for sub-talker sampling. |
| Max new tokens | 1 to 8192 | 200 | Maximum number of new codec tokens to generate. |
Every API field of Qwen 3 TTS 1.7B · Text to Speech is on its tool page: open its card above.
What can you do with Qwen3-TTS 1.7B in Claude Code?
Create spoken walkthroughs and help clips from the project's documentation. Use delivery instructions to compare how the same script sounds at different paces and emotional intensities.
Product tutorial narration
Turn a walkthrough script into measured speech, with delivery instructions suited to explaining each action.
Language listening exercises
Create spoken practice passages in supported languages. Start with a speaker whose native language matches the lesson.
Daily audio briefings
Read a prepared news or operations summary with a neutral tone and a consistent speaking pace.
Story narration
Give short stories a specified emotional delivery, then compare restrained and more expressive readings.
Multilingual explainer narration
Generate separate readings of reviewed translations in supported languages for educational or product videos.
Spoken help content
Create audio readings of common support answers with a patient tone and an unhurried pace.
Narration direction tests
Read the same script with different tone, pace, and emotion instructions before choosing a delivery for production.
More on Qwen3-TTS 1.7B and Claude Code
Overview
Qwen3-TTS 1.7B in other agents
Questions about Qwen3-TTS 1.7B in Claude Code
Can Claude Code use Qwen3-TTS 1.7B?
Yes. Add Spiral to Claude Code as an MCP server, and Claude Code can find and run Qwen3-TTS 1.7B like any built-in tool.
How much does Qwen3-TTS 1.7B cost in Claude Code?
Each item in `text` adds $0.117. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Claude Code plan is separate.
Do I need a fal account to use Qwen3-TTS 1.7B in Claude Code?
No. Spiral runs Qwen3-TTS 1.7B on its own fal account. You sign in to Spiral once and never paste a key.
How long does it take to set up Qwen3-TTS 1.7B in Claude Code?
About two minutes: add the hosted server, then sign in.
Does Claude Code wait while Qwen3-TTS 1.7B runs?
Qwen3-TTS 1.7B runs in the background. Claude Code starts the run, checks on it and gives you the audio when it's ready.
What is Qwen3-TTS 1.7B useful for?
It is a useful choice for narrated tutorials, spoken briefings, and lessons that need a specified tone or pace. You can direct the delivery with instructions such as calm, measured, or excited.
Which languages does Qwen3-TTS support?
It supports Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish, and Italian. Qwen recommends using a speaker's native language for the best quality.
How should I describe the delivery?
Specify tone, pace, and emotion in concrete terms. For example, try: Speak slowly with a reassuring tone and restrained emotion.
Can Qwen3-TTS generate long narration?
Qwen evaluated speech exceeding 10 minutes, but that result does not establish a maximum duration per request. Review long recordings for repeated words, omissions, and changes in delivery.
What audio sample rate does Qwen3-TTS use?
The native decoder output is 24 kHz. The 12Hz name refers to its speech tokenizer rather than the audio sample rate.
Start using Qwen3-TTS 1.7B in Claude Code
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Claude Code:
set up https://spiral.ai/SKILL.md


