Audio models handle different tasks: speech synthesis reads text aloud, transcription turns a recording into text, and music models create sound. Each card identifies the model’s task. Voices, languages, formats and duration limits depend on the selected tool.
Audio billing units differ: text characters, seconds of source audio or the duration of generated output. Compare prices together with their units; the request’s charge appears beside the result and in your history. Transcription takes a recording, speech synthesis takes text, and music generation uses a separate model.
Choose a model and send your request from the workspace. Its cost is calculated automatically and appears beside the result and in your history. Availability, capabilities and rates can change; check the selected model before sending a request.
35 models in this category · Updated:
ace-step-15Feature-rich song generation with optional lyrics and detailed musical controls.
tts-chatterbox-hdChatterbox HD is Resemble AI's open-weight text-to-speech model, known for expressive, natural delivery. It offers 9 voices and returns WAV audio.
elevenlabs-tts-multilingual-v2Multilingual text-to-speech using ElevenLabs. Supports 29 languages with high-quality natural-sounding voices, configurable speed, and accent accuracy.
elevenlabs-musicHigh-quality instrumental music generation with configurable duration. Best for polished, production-ready tracks across a wide range of genres.
elevenlabs-music-v2High-quality realistic music generation with optional vocals and configurable duration. Best for polished, production-ready tracks across a wide range of genres.
elevenlabs-music-v2-5Latest ElevenLabs music generation with higher-quality realistic tracks, optional vocals, and configurable duration. Best for polished, production-ready songs across a wide range of genres.
elevenlabs/scribe-v2ElevenLabs Scribe V2 is ElevenLabs' speech-to-text model, built for accurate transcription across many languages. It accepts a language hint and returns word-level timestamps with speaker labels.
elevenlabs-sound-effects-v2Generate high-quality sound effects from text descriptions using ElevenLabs. Ideal for films, games, and digital content with configurable duration.
elevenlabs-tts-v3Generate natural text-to-speech audio using ElevenLabs Eleven-v3. High-quality voices with stability control and automatic text normalization.
elevenlabs-tts-v4Generate expressive text-to-speech audio using ElevenLabs Eleven-v4. High-quality voices with audio tag support and automatic text normalization.
elevenlabs-tts-v4-turboGenerate fast text-to-speech audio using ElevenLabs Eleven-v4 Turbo. High-quality low-latency voices with audio tag support and automatic text normalization.
tts-elevenlabs-turbo-v2-5ElevenLabs Turbo v2.5 is ElevenLabs' low-latency multilingual speech model, balancing quality and speed for real-time use. It offers 21 voices and returns MP3 audio.
elevenlabs-voice-changerRe-record any speech in a different voice with ElevenLabs Voice Changer. Keeps the original delivery, pacing and emotion.
tts-gemini-3-1-flashGemini 3.1 Flash TTS is Google's Gemini-based speech model, which follows natural-language instructions for tone and pacing. It offers 30 voices and returns MP3, Opus or WAV audio.
tts-gradium-v1Gradium TTS is Gradium's text-to-speech model for natural, low-latency voice generation. It offers 12 voices and returns WAV, PCM or Opus audio.
tts-inworld-1-5-maxInworld TTS-1.5 Max is the higher-quality tier of Inworld's TTS-1.5 speech model, made for expressive delivery in games, companions and voice agents. It offers 14 voices and returns WAV audio.
tts-kokoroKokoro is hexgrad's open-weight 82M-parameter text-to-speech model. It speaks in 54 voices across several languages and returns MP3, Opus, AAC, FLAC, WAV or PCM audio.
lyria-3-proGoogle's Lyria 3 Pro generates full-length, structured songs up to 3 minutes long from a single text prompt. Supports vocals, lyrics, and multi-language generation across genres.
minimax-music-v2Full song generation with vocals and lyrics. Provide your own lyrics with verse/chorus structure for complete songs with singing.
minimax-music-v25Advanced song generation with vocals, lyrics optimizer, and instrumental mode. Supports structure tags and up to 3500 character lyrics.
minimax-music-v26Latest MiniMax song generation with vocals, instrumental mode, and support for rich structure tags in lyrics.
tts-minimax-speech-02-hdClone your voice from a short recording and generate natural speech in it across 30+ languages.
mmaudio-v2-text-to-audioGenerate synchronized audio and sound effects from text prompts with MMAudio V2.
tts-orpheusOrpheus TTS is Canopy Labs' open-weight speech model, built on Llama 3B for natural, emotive delivery. It offers 8 voices and returns WAV audio.
nvidia/parakeet-tdt-0.6b-v3Parakeet TDT 0.6B v3 is NVIDIA's open-weight speech recognition model, with 600 million parameters and support for 25 European languages, which it detects on its own. It transcribes quickly, with punctuation and capitalization.
tts-qwen3-0-6bQwen 3 TTS 0.6B is the smaller of Alibaba's two open-weight Qwen3-TTS models, built for fast, low-cost speech. It offers 9 voices and returns MP3 audio.
tts-qwen3-1-7bQwen 3 TTS 1.7B is the larger of Alibaba's two open-weight Qwen3-TTS models, trading some speed for higher-quality speech. It offers 9 voices and returns MP3 audio.
seed-audio-1-0Generate expressive multilingual speech and audio from a text prompt with BytePlus Seed Audio 1.0 (20 languages, timestamp length control).
sonilo-v1-1-musicGenerate licensed, commercial-use-safe music with precise control over style, mood, instrumentation, and duration.
sonilo-v1-1-sound-effectsGenerate licensed, commercial-use-safe sound effects with precise control over type, texture, intensity, and duration.
stable-audio-25Fast, lightweight audio generation for sound effects, ambient textures, and short musical clips. Flexible duration from 5 seconds to over 3 minutes.
openai/whisper-large-v3Whisper Large V3 is OpenAI's open-weight speech recognition model, trained on about 5 million hours of audio and able to transcribe around 100 languages. It's a robust general-purpose choice for multilingual transcription.
fal-ai/wizperWizper is an optimized deployment of OpenAI's Whisper Large V3, transcribing the same languages with faster turnaround. It accepts a language hint and returns segment timestamps.
stt-xai-v1xAI Speech to Text v1 is xAI's speech recognition model, which transcribes recorded audio into text. It accepts a language hint and returns word-level timestamps.
tts-xai-v1xAI TTS v1 is xAI's text-to-speech model, made for narration, assistants and voice apps. It offers 26 voices and returns MP3, WAV or PCM audio.
The UNCENSORED label comes from model specifications and identifies models designed for fewer refusals. It does not guarantee accuracy or an answer to every request. Max is a reasoning-effort setting on models that support it; each card lists the available modes. Rules and capabilities depend on the selected model.
OPEN THE INTERACTIVE CATALOG ↗