qwen3-tts-instruct-flash-realtime

Will be retired on October 10, 2026View announcement

Copy success!

Add to Compare

Overview

qwen3-TTS-Flash model is Tongyi's latest real-time speech synthesis model. The Instruct model processes the synthesis effect through natural language, ensuring highly appropriate emotional and expressive speech in different contexts. Currently, it supports 25 timbres for both Chinese and English Instruct adjustments. This model is a snapshot version from January 22, 2026.

Input

Text

Output

Audio

Features

Prefix Completion

Enable Partial Mode when calling the Qwen API to make the model continue strictly from your provided prefix text.View docs

Function Calling

Use function calling to connect large language models with external tools and systems.View docs

Cache

Context Cache stores shared prefixes for long-context requests to reduce repeated computation, improve latency, and lower cost.View docs

Structured Outputs

Structured Outputs help ensure the model returns a JSON string in the expected format.View docs

Batches

Asynchronously process requests in batches to reduce costs.View docs

Web Search

Enable web search so the model can answer with real-time retrieved data.View docs

Fine-tuning

Train models on sample data to better adapt them to specific tasks.View docs

Pricing

TTS
$0.143Per 10,000 characters

Rate Limits

RPMRequests Per Minute
180

API Reference

Call API

Copy success!

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465666768697071727374757677787980818283848586878889909192939495

import os
import base64
import threading
import time
import dashscope
from dashscope.audio.qwen_tts_realtime import *

qwen_tts_realtime: QwenTtsRealtime = None
text_to_synthesize = [
    'Right?~ I especially love this kind of supermarket，',
    'Especially during the New Year',
    'Going to the supermarket',
    'It just makes me feel',
    'Super, super happy!',
    'I want to buy so many things!'
]

DO_VIDEO_TEST = False

def init_dashscope_api_key():
    """
        Set your DashScope API-key. More information:
        https://github.com/aliyun/alibabacloud-bailian-speech-demo/blob/master/PREREQUISITES.md
    """

    if 'DASHSCOPE_API_KEY' in os.environ:
        dashscope.api_key = os.environ[
            'DASHSCOPE_API_KEY']  # load API-key from environment variable DASHSCOPE_API_KEY
    else:
        dashscope.api_key = 'your-dashscope-api-key'  # set API-key manually



class MyCallback(QwenTtsRealtimeCallback):
    def __init__(self):
        self.complete_event = threading.Event()
        self.file = open('result_24k.pcm', 'wb')

    def on_open(self) -> None:
        print('connection opened, init player')

    def on_close(self, close_status_code, close_msg) -> None:
        self.file.close()
        print('connection closed with code: {}, msg: {}, destroy player'.format(close_status_code, close_msg))

    def on_event(self, response: str) -> None:
        try:
            global qwen_tts_realtime
            type = response['type']
            if 'session.created' == type:
                print('start session: {}'.format(response['session']['id']))
            if 'response.audio.delta' == type:
                recv_audio_b64 = response['delta']
                self.file.write(base64.b64decode(recv_audio_b64))
            if 'response.done' == type:
                print(f'response {qwen_tts_realtime.get_last_response_id()} done')
            if 'session.finished' == type:
                print('session finished')
                self.complete_event.set()
        except Exception as e:
            print('[Error] {}'.format(e))
            return

    def wait_for_finished(self):
        self.complete_event.wait()


if __name__  == '__main__':
    init_dashscope_api_key()

    print('Initializing ...')

    callback = MyCallback()

    qwen_tts_realtime = QwenTtsRealtime(
        model='qwen3-tts-instruct-flash-realtime-2026-01-22',
        callback=callback, 
        )

    qwen_tts_realtime.connect()
    qwen_tts_realtime.update_session(
        voice = 'Cherry',
        response_format = AudioFormat.PCM_24000HZ_MONO_16BIT,
        mode = 'server_commit'        
    )
    for text_chunk in text_to_synthesize:
        print(f'send texd: {text_chunk}')
        qwen_tts_realtime.append_text(text_chunk)
        time.sleep(0.1)
    qwen_tts_realtime.finish()
    callback.wait_for_finished()
    print('[Metric] session: {}, first audio delay: {}'.format(
                    qwen_tts_realtime.get_session_id(), 
                    qwen_tts_realtime.get_first_audio_delay(),
                    ))