Spiral gives your agent every tool with one account. Browse the tools →

Codex · AudioReleased 2026-01-22

How to Use Qwen3 TTS 0.6B in Codex

Use Qwen3 TTS 0.6B for multilingual narration with preset voices. Unlike Qwen3 TTS 1.7B CustomVoice, its reference implementation ignores instructions that describe a speaking style.

Prompts to try with Qwen3 TTS 0.6B in Codex

Copy an example into Codex, or ask in your own words.

Use Qwen3 TTS 0.6B to narrate an 80-word English introduction based on this project's README, using Aiden. Save the script and audio together in the project's narration directory.

Qwen3 TTS 0.6BCodex
Try asking
  • Read the onboarding documentation and prepare one short English narration script per task. Use Qwen3 TTS 0.6B with Ryan to generate a separate recording for each script.

  • Use the approved Japanese tutorial text in this project to create narration with Qwen3 TTS 0.6B and Ono_Anna. Preserve the wording and generate one clip per section.

  • Create two 60-word English introductions to the demo from the project's feature documentation. Generate both with Qwen3 TTS 0.6B using Aiden so I can compare the wording by listening.

  • Use Qwen3 TTS 0.6B with Ryan to generate three readings of the project's English pronunciation test script at temperatures 0.7, 0.8, and 0.9. Keep all other settings fixed and label the recordings with the temperature.

How to set up Qwen3 TTS 0.6B in Codex

Pick the way that suits you. Each takes about two minutes.

Use Spiral's MCP server

  1. Add the hosted server
    codex mcp add spiral --url https://api.spiral.ai/mcp
  2. Sign in

    Codex opens the browser: sign in to Spiral and click Authorize. No API key. codex mcp list then shows Spiral.

    codex mcp login spiral

Use the Spiral skill

  1. Send Codex this message

    Your agent reads Spiral's skill and sets itself up.

    set up https://spiral.ai/SKILL.md
  2. Sign in or paste a key

    Approve Spiral in the browser, or give the agent a key from the console.

Use the Spiral CLI

  1. 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

  1. Create an API key

    In the console, under API keys.

  2. Run Qwen3 TTS 0.6B

    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/0.6b",
        "input": {
          "text": "I feel like I'\''m taking crazy pills! How can something be both a square and a circle at the same time? It defies all logic!"
        }
      }'

Then paste one of the prompts above: Codex finds Qwen3 TTS 0.6B on Spiral, shows the price and runs it.

Qwen3 TTS 0.6B providers and prices

Spiral runs Qwen3 TTS 0.6B through fal. Open a card for today's price, with every fee included.

falAudio
Qwen 3 TTS 0.6B · Text to Speech
fal-ai/qwen-3-tts/text-to-speech/0.6b

Settings you can ask for

Qwen3 TTS 0.6B works from your prompt. Ask Codex for any of these in your prompt.

SettingAllowed valuesDefaultDescription
VoiceVivian, 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.

LanguageAuto, English, Chinese, Spanish, French, German, Italian, Japanese and 3 moreAuto

The language of the voice.

Top ka whole number from 050

Top-k sampling parameter.

Top p0 to 11

Top-p sampling parameter.

Temperaturea number up to 10.9

Sampling temperature; higher => more random.

Repetition penaltya number from 01.05

Penalty to reduce repeated tokens/codes.

Subtalker dosampleon or offon

Sampling switch for the sub-talker.

Subtalker top ka whole number from 050

Top-k for sub-talker sampling.

Subtalker top p0 to 11

Top-p for sub-talker sampling.

Subtalker temperature0 to 10.9

Temperature for sub-talker sampling.

Max new tokens1 to 8192200

Maximum number of new codec tokens to generate.

Every API field of Qwen 3 TTS 0.6B · Text to Speech is on its tool page: open its card above.

What can you do with Qwen3 TTS 0.6B in Codex?

Create spoken tutorials and product walkthroughs from your project's documentation. A coding agent can prepare narration scripts, generate clips, and keep the text associated with each recording.

Product walkthrough narration

Turn an approved walkthrough script into spoken explanations of each product task. Generate separate clips for sections that may change later.

Tutorial audio

Create narration for lessons and software tutorials from finished instructional text.

Localized announcements

Generate spoken announcements from approved translations in supported languages. Review each language before publishing.

Language listening exercises

Convert practice sentences into individual audio clips for vocabulary review, listening comprehension, and pronunciation comparison.

Spoken daily briefings

Have an agent prepare a concise written digest, then generate a spoken edition using the same preset voice each day.

Help article narration

Make spoken editions of help articles after rewriting headings, links, and interface references into sentences that work aloud.

Phone menu recordings

Generate individual recordings for greetings, department choices, and business hours from approved scripts.

Narration script comparisons

Read alternate introductions or explanations with the same voice so editors can compare wording by listening.

Questions about Qwen3 TTS 0.6B in Codex

Can Codex use Qwen3 TTS 0.6B?

Yes. Add Spiral to Codex as an MCP server, and Codex can find and run Qwen3 TTS 0.6B like any built-in tool.

How much does Qwen3 TTS 0.6B cost in Codex?

Each item in `text` adds $0.091. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Codex plan is separate.

Do I need a fal account to use Qwen3 TTS 0.6B in Codex?

No. Spiral runs Qwen3 TTS 0.6B 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 0.6B in Codex?

About two minutes: add the hosted server, then sign in.

Does Codex wait while Qwen3 TTS 0.6B runs?

Qwen3 TTS 0.6B runs in the background. Codex starts the run, checks on it and gives you the audio when it's ready.

What is Qwen3 TTS 0.6B useful for?

Use it to turn written scripts into speech for tutorials, product walkthroughs, language exercises, and spoken updates. Its preset voices make it a practical choice when you already have the words to read.

Which languages does Qwen3 TTS 0.6B support?

It supports Chinese, English, Japanese, Korean, German, French, Russian, Portuguese, Spanish, and Italian. Prepare the script in the language you want spoken.

Can I tell Qwen3 TTS 0.6B to whisper or sound excited?

Do not rely on those instructions: Qwen's reference implementation disables style instructions for 0.6B, despite broader claims in its model card. Qwen3 TTS 1.7B CustomVoice supports instruction-based style control.

How should I choose a voice?

Start with a voice whose native language matches your script, as Qwen recommends. Compare a short passage before generating the full narration.

What audio sample rate does Qwen3 TTS 0.6B produce?

Its 12Hz tokenizer specifies a native output sample rate of 24 kHz. Use that as the source rate when planning audio editing or resampling.

Get started

Start using Qwen3 TTS 0.6B in Codex

One Spiral account for every tool your agent needs. No subscription; you pay per run.

Paste this into Codex:

Set up Codex

set up https://spiral.ai/SKILL.md
Are you an AI agent? See setup options