Gemini 3.8 Flash TTS vs Qwen Audio 3.0 TTS

Gemini 3.8 Flash TTS vs Qwen Audio 3.0 TTS at a glance

Gemini 3.8 Flash TTS from Google and Qwen Audio 3.0 TTS from Alibaba Cloud are both in the Audio Generation category, available through the EmpirioLabs API.

Key differences

• Gemini 3.8 Flash TTS is the newer release (Sep 22, 2026); Qwen Audio 3.0 TTS was released on Jul 20, 2026.
• Capabilities listed only for Gemini 3.8 Flash TTS: multi speaker, voice control, streaming.
• Capabilities listed only for Qwen Audio 3.0 TTS: expressive prosody, emotion control, dialect, low latency.

Gemini 3.8 Flash TTS vs Qwen Audio 3.0 TTS: specs

Gemini 3.8 Flash TTS vs Qwen Audio 3.0 TTS: API pricing

Live rates from the EmpirioLabs catalog for both models. Each rate card always matches what the API charges.

Gemini 3.8 Flash TTS vs Qwen Audio 3.0 TTS: API parameters

Request parameters each model lists on EmpirioLabs, with the allowed values or range where one applies. A dash marks a parameter the model does not list.

Gemini 3.8 Flash TTS vs Qwen Audio 3.0 TTS: limits and notes

Gemini 3.8 Flash TTS

Limits

• Text: up to 5,000 characters per request
• Audio billing: 32 output tokens per second of generated audio
• 130 languages. The language of the text is detected automatically.

Voices

• The 30 voices in the voice list speak every supported language.
• GET /v1/voices?model=gemini-3-8-flash-tts lists the full library of more than 2,000 voices across 30 locales, with the language, accent, gender, pitch, and a description of each. Filter it with language, gender, pitch, or q. Any id it returns works as voice or voice2; when you also set language, use the voice's own language.

Custom voices

• Set voice to custom and pass two recordings of the same adult speaker as https URLs: voice_audio_url, 10 to 30 seconds of natural speech, and voice_consent_audio_url, the speaker reading "I am the owner of this voice and I consent to Google using this voice to create a synthetic voice model." The statement is accepted in 30 languages; consent_statements on that parameter lists each wording.
• Both must be recordings of a real person, in a quiet room on the same microphone. AI-generated audio is not accepted. WAV, MP3, M4A, OGG, WEBM and FLAC are accepted, up to 15 MB each.
• The response includes voice_id and voice_id_expires_at. Pass the voice_id as voice or voice2 to reuse the voice for 7 days without new recordings. A voice_id works only for the account that created it.
• This model keeps no list of custom voices, so keep the voice_id from the response. A custom voice ends when its id expires, so there is nothing to delete.
• In the Playground, set Voice to Custom to use a voice made in the same tab again, or copy its id.

Delivery direction

• Put direction for the whole passage in style_prompt rather than in the text, for example "warm, unhurried, and reassuring".
• Place vocal tags inline where the sound should happen. Use these English tags even when the text is in another language:

<argh> <breath> <heavy breath> <exhales> <cackle> <cheer> <chuckle> <chuckles> <cough> <cry> <gasp> <giggle> <groan> <growl> <grunt> <grr> <hiss> <laugh> <laughter> <moan> <pant> <pff> <phew> <scream> <shout> <shriek> <sigh> <sighs> <sneeze> <snicker> <snort> <sob> <throat-clearing> <tsk> <whimper> <whispers> <whispering> <yawn> <short pause> <long pause>

• Capitalize a word to stress it. Wrap a listener reaction in pipes, for example |mm-hmm|, to add it without starting a new turn.

Multi-speaker

• Up to two speakers per request. Start each line with a speaker name and a colon, for example Speaker1: and Speaker2:. The names must match speaker1_name and speaker2_name.

Output

• WAV, MP3, OGG (Opus), ALAW, or MULAW, at 8,000 to 48,000 Hz.
• POST /v1/audio/speech:stream sends 16-bit PCM chunks at 24,000 Hz while the audio is generated, then the complete file in the requested format. speed is available on POST /v1/audio/speech only.

Qwen Audio 3.0 TTS

Tiers

• Plus: highest audio quality and expressiveness, for content creation, audiobooks, dubbing, and brand voice work
• Flash: tuned for real-time interaction with lower first-packet latency, for voice assistants and live agents
• Switch with the model_tier parameter, or send qwen-audio-3.0-tts-plus or qwen-audio-3.0-tts-flash as the model id

Voices

• Over 1,000 base voices, all available on both tiers, so changing tier keeps your selected voice
• Six flagship system voices are tier specific: longanlingxin and longanlufeng on Plus, and longanhuan_v3.6, longjielidou_v3.6, loongeva_v3.6, loongjohn on Flash
• Browse the full catalog with GET /v1/voices, and pass any voice through voice_id

Custom voices

• Set voice to custom and pass voice_audio_url, an https URL to 10 to 20 seconds of clear speech from one speaker.
• WAV, MP3, M4A, OGG, WEBM and FLAC are accepted, up to 15 MB. Record in a quiet room, without music or other voices.
• The voice is made for the selected model_tier. The response includes voice_id and voice_id_expires_at: pass the voice_id as voice or voice_id, with the same model_tier, to reuse the voice for 7 days. Only the account that created it can use it.
• Each account can hold 10 custom voices on this model at a time.
• GET /v1/voices?model=qwen-audio-3-0-tts lists the model's voices, including your custom voices with their ids. Add type=custom to list custom voices alone.
• DELETE /v1/voices/{voice_id}?model=qwen-audio-3-0-tts deletes a custom voice, which frees its place at once.
• In the Playground, set Voice to Custom to use a saved voice again, copy its id, or delete it.
• Creating, listing and deleting custom voices adds no charge.

Input limits

• Up to 20,000 characters per request
• instruction accepts up to 128 characters
• SSML input is not supported. Use instruction for delivery direction, and inline tags such as [excited] or [whispers] inside the text

Billing

• Billed per character of input text at the selected tier's rate

How to call Gemini 3.8 Flash TTS and Qwen Audio 3.0 TTS

Gemini 3.8 Flash TTS serves speech through POST /v1/audio/speech and returns playable audio. Send the text to speak as input with the model id gemini-3-8-flash-tts. Both models are called through the same endpoint. Set model to gemini-3-8-flash-tts or qwen-audio-3-0-tts. Get an API key from the EmpirioLabs dashboard.

curl https://api.empiriolabs.ai/v1/audio/speech \
  -H "Authorization: Bearer $EMPIRIOLABS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gemini-3-8-flash-tts",
    "input": "Welcome to EmpirioLabs. Your build just finished."
  }' \
  --output speech.mp3

Gemini 3.8 Flash TTS vs Qwen Audio 3.0 TTS: tests and guides

Prompt tests and comparison runs that cover both models.

Guides Qwen Audio 3.1 TTS vs Qwen Audio 3.0 TTS vs Gemini 3.8 Flash TTS vs StepAudio 3 TTS: one script We gave Qwen Audio 3.1 TTS, Qwen Audio 3.0 TTS, Gemini 3.8 Flash TTS and StepAudio 3 TTS the same script and delivery...

Gemini 3.8 Flash TTS vs Qwen Audio 3.0 TTS: common questions

What is the difference between Gemini 3.8 Flash TTS and Qwen Audio 3.0 TTS?

Gemini 3.8 Flash TTS is the newer release (Sep 22, 2026); Qwen Audio 3.0 TTS was released on Jul 20, 2026. Capabilities listed only for Gemini 3.8 Flash TTS: multi speaker, voice control, streaming. Capabilities listed only for Qwen Audio 3.0 TTS: expressive prosody, emotion control, dialect, low latency. The specs, rate cards, and parameters on this page come from the live EmpirioLabs catalog.

Which is newer, Gemini 3.8 Flash TTS or Qwen Audio 3.0 TTS?

Gemini 3.8 Flash TTS is the newer release (Sep 22, 2026); Qwen Audio 3.0 TTS was released on Jul 20, 2026.

Which has the lower rate, Gemini 3.8 Flash TTS or Qwen Audio 3.0 TTS?

The two models are priced on different units or tiers, so compare the live rate cards on this page. They always match what the API charges.

Can I switch between Gemini 3.8 Flash TTS and Qwen Audio 3.0 TTS without rebuilding my integration?

Yes. Both models use POST /v1/audio/speech with the same EmpirioLabs API key, so most integrations switch by changing the model value. Check the parameters table on this page first if you rely on a model-specific parameter.

Can I try Gemini 3.8 Flash TTS and Qwen Audio 3.0 TTS in the browser before integrating?

Yes. Open each model in the EmpirioLabs playground and run the same prompt on both, with the parameters the API exposes.