Qwen 3 TTS - Text to Speech [1.7B]

alibaba/qwen-3-tts/1.7b

Qwen 3 Tts 1.7b is Alibaba's text-to-speech AI model. Generate human-like voiceovers with expressive intonation, multilingual support, and customizable voice characteristics.

Base price
$0.0001USD / run
Execution
async
Model type
audio
Input fields
15
Production route

Try the model

Playground

Open playground
Input
Optional prompt to guide the style of the generated speech. This prompt will be ignored if a speaker embedding is provided.
The text to be converted to speech.
01
Temperature for sub-talker sampling. Range: 0 to 1.
URL to a speaker embedding file in safetensors format, from `fal-ai/qwen-3-tts/clone-voice` endpoint. If provided, the TTS model will use the cloned voice for synthesis instead of the predefined voices.
Penalty to reduce repeated tokens/codes. Minimum: 0.
The language of the voice. Allowed values: Auto, English, Chinese, Spanish, French, German, Italian, Japanese, Korean, Portuguese, Russian.
01
Top-p for sub-talker sampling. Range: 0 to 1.
01
Top-p sampling parameter. Range: 0 to 1.
Optional reference text that was used when creating the speaker embedding. Providing this can improve synthesis quality when using a cloned voice.
Top-k sampling parameter. Minimum: 0.
Top-k for sub-talker sampling. Minimum: 0.
01
Sampling temperature; higher => more random. Range: 0 to 1.
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. Allowed values: Vivian, Serena, Uncle_Fu, Dylan, Eric, Ryan, Aiden, Ono_Anna, Sohee.
18192
Maximum number of new codec tokens to generate. Range: 1 to 8192.
Sampling switch for the sub-talker.
OutputExample
  1. What would you like to do next?

    Open original

Examples

Select an example to load its inputs and result into the playground.

Model card

Specifications

Pricing

Base price
$0.0001 / run

Context & modalities

Input
Schema-defined
Output
audio

Capabilities

Chat
Not supported
Vision
Not supported
Reasoning
Not supported
Structured output
Not supported
Function calling
Not supported
Audio input
Not supported

Access

Provider
Alibaba
Model ID
alibaba/qwen-3-tts/1.7b
Execution
async
Base URL
https://api.sandbase.ai
API
Unified Run API
Endpoint
/v1/run

API README

Qwen 3 TTS - Text to Speech [1.7B]

Qwen 3 TTS 1.7B, the open-weights Qwen speech model: text to speech in 10 languages with nine preset voices, or a cloned voice from a speaker-embedding file. The 1.7B model gives more natural prosody than 0.6B.

Fields

FieldTypeDefaultNotes
promptstring, required—Text to speak.
textstring—Alias of prompt; if both are sent, text is used.
subtalker_temperaturenumber 0–10.9Sub-model temperature.
speaker_voice_embedding_file_urlstring—Optional cloned-voice embedding (safetensors) to use instead of a preset voice.
repetition_penaltynumber1.05Reduces repeated sounds.
languageAuto / English / Chinese / Spanish / French / German / Italian / Japanese / Korean / Portuguese / RussianAutoSpoken language; Auto detects it.
subtalker_top_pnumber 0–11Sub-model sampling setting.
top_pnumber 0–11Sampling setting.
reference_textstring—Transcript used when the embedding was created; improves cloned-voice quality.
top_kinteger50Sampling setting; leave at default unless delivery is unstable.
subtalker_top_kinteger50Sub-model sampling setting.
temperaturenumber 0–10.9Higher values give more varied delivery.
voiceVivian / Serena / Uncle_Fu / Dylan / Eric / Ryan / Aiden / Ono_Anna / Sohee—Preset voice; ignored when a speaker embedding is supplied.
max_new_tokensinteger 1–8192200Caps the generated audio length; raise it for long text.
subtalker_dosamplebooleantrueSampling switch for the acoustic sub-model.

Billing

$0.09 per 1,000 characters of input text, billed per character.

TextPrice
100 characters$0.009
1,000 characters$0.09
5,000 characters$0.45

Failed runs are not charged.

Prompting

  • Use punctuation to control pauses; break long passages into paragraphs.
  • Pick a voice whose primary language matches the text.

Limits

  • max_new_tokens (default 200) caps the audio length; long text needs a higher value.

Also on SandBase

Related: alibaba/qwen-3-tts/0.6b.

Start building

Send your first request

OpenAI-compatible endpoint with unified authentication and usage tracking.

Production API
Unified Run API endpoint
https://api.sandbase.ai/v1/run
Model ID
alibaba/qwen-3-tts/1.7b
# The quoted heredoc keeps Unicode and shell metacharacters unchanged.
result=$(curl --fail-with-body --silent \
  -X POST "https://api.sandbase.ai/v1/run" \
  -H "Authorization: Bearer $SANDBASE_API_KEY" \
  -H "Content-Type: application/json" \
  --data-binary @- <<'SANDBASE_JSON'
{
  "model": "alibaba/qwen-3-tts/1.7b",
  "text": "I am solving the equation: x = [-b ± √(b²-4ac)] / 2a? Nobody can — it's a disaster (◍•͈⌔•͈◍), very sad!",
  "top_k": 50,
  "top_p": 1,
  "voice": "Vivian",
  "prompt": "Very happy.",
  "language": "English",
  "temperature": 0.9,
  "max_new_tokens": 200,
  "reference_text": "Okay. Yeah. I resent you. I love you. I respect you. But you know what? You blew it! And it is all thanks to you.",
  "subtalker_top_k": 50,
  "subtalker_top_p": 1,
  "repetition_penalty": 1.05,
  "subtalker_dosample": true,
  "subtalker_temperature": 0.9,
  "speaker_voice_embedding_file_url": "https://static.sandbase.ai/examples/alibaba/qwen-3-tts/1.7b/output_url_0.safetensors"
}
SANDBASE_JSON
)
run_id=$(printf '%s' "$result" | jq -r .id)
for attempt in $(seq 1 120); do
  status=$(printf '%s' "$result" | jq -r .status)
  case "$status" in completed|failed|timeout) break ;; esac
  sleep 2
  result=$(curl --fail-with-body --silent \
    -H "Authorization: Bearer $SANDBASE_API_KEY" \
    "https://api.sandbase.ai/v1/run/$run_id")
done
status=$(printf '%s' "$result" | jq -r .status)
[ "$status" = completed ] || { echo "Generation ended: $status" >&2; exit 1; }
printf '%s\n' "$result"

Choose your model

Compare models

All image & video models
ModelTypeDefault priceReleased
AlibabaAudio$0.00 / runJan 26, 2026
AlibabaAudio$0.00 / runJan 26, 2026

Questions

FAQ

How do I call Alibaba Qwen 3 TTS - Text to Speech [1.7B] through SandBase?

Create a SandBase API key, then send requests with the model ID "alibaba/qwen-3-tts/1.7b" to Unified Run API (/v1/run) at https://api.sandbase.ai. The request examples on this page show the exact payload.

Do I need a separate Alibaba account?

No. One SandBase API key and balance gives you access to Alibaba Qwen 3 TTS - Text to Speech [1.7B] and the other models in the catalog; you do not need to sign up with Alibaba separately.