Skip to content
SF·SIX
SF·SIX / SOURCE-BACKED RESEARCH

Audio generation models

Browse models that generate audio. Check whether an endpoint is intended for speech, music or another audio task.

117 models with current provider prices, most widely served first

Checked · Sources: OpenRouter, fal.ai · Methodology

Models

Lyria 3 Pro Preview

Google DeepMind · text, audio

From $0.00 in / $0.00 out per 1M tokens · observed 2026-10-17

1,048,576 context tokens · 1 current source-qualified record

Lyria 3 Clip Preview

Google DeepMind · text, audio

From $0.00 in / $0.00 out per 1M tokens · observed 2026-10-17

1,048,576 context tokens · 1 current source-qualified record

GPT Audio

OpenAI · text, audio

From $2.50 in / $10.00 out per 1M tokens · observed 2026-10-17

128,000 context tokens · 1 current source-qualified record

GPT Audio Mini

OpenAI · text, audio

From $0.60 in / $2.40 out per 1M tokens · observed 2026-10-17

128,000 context tokens · 1 current source-qualified record

Mirelo SFX2.0 (video-to-audio)

Mirelo AI · audio

1 current source-qualified record

Mirelo SFX2.0 (text-to-audio)

Mirelo AI · audio

1 current source-qualified record

Kotoba Text-to-Speech

kotoba-technologies · audio

1 current source-qualified record

Kotoba Realtime Speech Translation

kotoba-technologies · audio

1 current source-qualified record

Eleven v4 Turbo

ElevenLabs · audio

1 current source-qualified record

Eleven v4

ElevenLabs · audio

1 current source-qualified record

Gemini 3.8 Flash TTS

Google DeepMind · audio

1 current source-qualified record

Gemini 3.8 Flash Lite TTS

Google DeepMind · audio

1 current source-qualified record

Lyria 3.5

Google DeepMind · audio

1 current source-qualified record

Mureka 9.5 Song Generation

Mureka · audio

1 current source-qualified record

Mureka 9.5 Podcast Generation

Mureka · audio

1 current source-qualified record

Mureka 9.5 Instrumental Generation

Mureka · audio

1 current source-qualified record

Elevenlabs Music v2.5

ElevenLabs · audio

1 current source-qualified record

Elevenlabs Music v2

ElevenLabs · audio

1 current source-qualified record

Grok Voice (realtime)

xAI · audio

1 current source-qualified record

Grok Voice

xAI · audio

1 current source-qualified record

MiniMax Music 3

MiniMax · audio

1 current source-qualified record

Qwen Audio 3.0 TTS (Flash)

Qwen (Alibaba) · audio

1 current source-qualified record

V1.1 Video to Sound Effects

sonilo · audio

1 current source-qualified record

V1.1 Text to Sound Effects

sonilo · audio

1 current source-qualified record

Seed Audio 1.0

ByteDance Seed · audio

1 current source-qualified record

Async Text to Speech Pro V1.0

async · audio

1 current source-qualified record

Zonos2 Text to Speech

Unattributed (via fal) · audio

1 current source-qualified record

Bytedance Seed Speech Text to Speech

ByteDance Seed · audio

1 current source-qualified record

V1.1

sonilo · audio

1 current source-qualified record

Sonilo V1.1 Text to Music

sonilo · audio

1 current source-qualified record

Stable Audio 3 Small SFX Base Audio Outpainting

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Small SFX Audio Outpainting

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Small SFX Base Audio Inpainting

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Small SFX Audio Inpainting

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Small SFX Base Audio to Audio

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 (small / sfx / audio-to-audio)

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Small SFX Base Text to Audio

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Small SFX Text to Audio

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Small Music Base Audio Outpainting

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Small Music Audio Outpainting

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Small Music Base Audio Inpainting

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Small Music Audio Inpainting

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Small Music Base Audio to Audio

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 (small / music / audio-to-audio)

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 (small / music / base / text-to-audio)

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Small Music Text to Audio

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Medium Base Audio Outpainting

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Medium Audio Outpainting

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Medium Base Audio Inpainting

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Medium Audio Inpainting

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Medium Base Audio to Audio

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Medium Audio to Audio

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 Medium Base Text to Audio

Unattributed (via fal) · audio

1 current source-qualified record

Stable Audio 3 (medium / text-to-audio)

Unattributed (via fal) · audio

1 current source-qualified record

Gemini 3.1 Flash Tts

Unattributed (via fal) · audio

1 current source-qualified record

Minimax Music 2.5

MiniMax · audio

1 current source-qualified record

Minimax Music 2.6

MiniMax · audio

1 current source-qualified record

Tada

Unattributed (via fal) · audio

1 current source-qualified record

Tada TTS 1B

Unattributed (via fal) · audio

1 current source-qualified record

Personaplex (realtime)

Unattributed (via fal) · audio

1 current source-qualified record

Personaplex

Unattributed (via fal) · audio

1 current source-qualified record

MiniMax Speech 2.8 [HD]

MiniMax · audio

1 current source-qualified record

MiniMax Speech 2.8 [Turbo]

MiniMax · audio

1 current source-qualified record

Qwen 3 TTS - Voice Design [1.7B]

Qwen (Alibaba) · audio

1 current source-qualified record

Qwen 3 TTS - Text to Speech [1.7B]

Qwen (Alibaba) · audio

1 current source-qualified record

Qwen 3 TTS - Text to Speech [0.6B]

Qwen (Alibaba) · audio

1 current source-qualified record

Qwen 3 TTS - Clone Voice [1.7B]

Qwen (Alibaba) · audio

1 current source-qualified record

Qwen 3 TTS - Clone Voice [0.6B]

Qwen (Alibaba) · audio

1 current source-qualified record

Sam Audio (separate)

Unattributed (via fal) · audio

1 current source-qualified record

Sam Audio (span-separate)

Unattributed (via fal) · audio

1 current source-qualified record

Sam Audio (visual-separate)

Unattributed (via fal) · audio

1 current source-qualified record

Elevenlabs Music

ElevenLabs · audio

1 current source-qualified record

Maya (batch)

Unattributed (via fal) · audio

1 current source-qualified record

Maya (stream)

Unattributed (via fal) · audio

1 current source-qualified record

Maya1

Unattributed (via fal) · audio

1 current source-qualified record

Minimax Music

MiniMax · audio

1 current source-qualified record

MiniMax Speech 2.6 [Turbo]

MiniMax · audio

1 current source-qualified record

MiniMax Speech 2.6 [HD]

MiniMax · audio

1 current source-qualified record

Audio Understanding

Unattributed (via fal) · audio

1 current source-qualified record

Kling TTS

Kuaishou · audio

1 current source-qualified record

MiniMax (Hailuo AI) Music v1.5

MiniMax · audio

1 current source-qualified record

Stable Audio 2.5 (audio-to-audio)

Stability AI · audio

1 current source-qualified record

Stable Audio 2.5 (text-to-audio)

Stability AI · audio

1 current source-qualified record

Stable Audio 25

Stability AI · audio

1 current source-qualified record

Chatterbox (text-to-speech / multilingual)

Resemble AI · audio

1 current source-qualified record

Minimax (preview / speech-2.5-hd)

MiniMax · audio

1 current source-qualified record

Minimax (preview / speech-2.5-turbo)

MiniMax · audio

1 current source-qualified record

MiniMax Voice Design

MiniMax · audio

1 current source-qualified record

Chatterboxhd (speech-to-speech)

Resemble AI · audio

1 current source-qualified record

Chatterboxhd (text-to-speech)

Resemble AI · audio

1 current source-qualified record

Chatterbox (speech-to-speech)

Resemble AI · audio

1 current source-qualified record

Chatterbox (text-to-speech)

Resemble AI · audio

1 current source-qualified record

Lyria2

Google DeepMind · audio

1 current source-qualified record

ACE Step Audio Outpaint

ACE Studio · audio

1 current source-qualified record

MiniMax Voice Cloning

MiniMax · audio

1 current source-qualified record

MiniMax Speech-02 Turbo

MiniMax · audio

1 current source-qualified record

MiniMax Speech-02 HD

MiniMax · audio

1 current source-qualified record

Dia Tts

Unattributed (via fal) · audio

1 current source-qualified record

Dia

Unattributed (via fal) · audio

1 current source-qualified record

Sound Effects Generator

CassetteAI · audio

1 current source-qualified record

Orpheus TTS

Unattributed (via fal) · audio

1 current source-qualified record

music generator

CassetteAI · audio

1 current source-qualified record

CSM-1B

Unattributed (via fal) · audio

1 current source-qualified record

DiffRhythm: Lyrics to Song

Unattributed (via fal) · audio

1 current source-qualified record

ElevenLabs TTS Multilingual v2

ElevenLabs · audio

1 current source-qualified record

Kokoro TTS

Kokoro · audio

1 current source-qualified record

Kokoro TTS (Brazilian Portuguese)

Kokoro · audio

1 current source-qualified record

Kokoro TTS (British English)

Kokoro · audio

1 current source-qualified record

Kokoro TTS (French)

Kokoro · audio

1 current source-qualified record

Kokoro TTS (Hindi)

Kokoro · audio

1 current source-qualified record

Kokoro TTS (Italian)

Kokoro · audio

1 current source-qualified record

Kokoro TTS (Japanese)

Kokoro · audio

1 current source-qualified record

Kokoro TTS (Mandarin Chinese)

Kokoro · audio

1 current source-qualified record

Kokoro TTS (Spanish)

Kokoro · audio

1 current source-qualified record

Zonos-Audio-Clone

Unattributed (via fal) · audio

1 current source-qualified record

MMAudio V2 Text to Audio

MMAudio · audio

1 current source-qualified record

MiniMax (Hailuo AI) Music

MiniMax · audio

1 current source-qualified record

Capabilities are source-reported and do not constitute an independent quality assessment. Community catalog references for this capability are listed in the interactive directory.