Speech models, tools and agents
386 in the catalogue today. Every one with what it costs, who sells it and where it stands. 386 in all, a hundred to a page; this is page 4 of 4.
Models that read text aloud. They are usually priced per character, which makes the arithmetic simple: a thousand characters is roughly a paragraph. The real choices are the voice itself, whether you may clone one, how much control you have over emotion and pacing, and latency — a model that sounds wonderful in a rendered file may be too slow to hold a conversation.
- indic-speakBodhan AI
- inworld-tts-1.5-maxInworld AI16th of 41$0.00005per character1 selling
- inworld-tts-1.5-miniInworld AI$0.000025per character1 selling
- kan-bayashi_ljspeech_vitsESPnet
- kani-tts-2-ennineninesix
- kani-tts-2-ptnineninesix
- kani-tts-370mnineninesix
- kitten-tts-2Kitten ML
- kitten-tts-nano-0.8-fp32Kitten ML
- kugelaudio-0-openKugelaudio
- magenta-realtime-2Google
- magpie_tts_multilingual_357mNVIDIA
- marvis-tts-250m-v0.1MARVIS
- maya2-nativeMaya Research
- metavoice-1B-v0.1MetaVoice
- metricgan-plus-voicebankSpeechBrain
- midashenglm-genSpeech Team, Xiaomi MiLM Plus
- mini-omnigpt-omni
- mms-ttsMeta
- mms-tts-engMeta
- mustangoDeep Cognition and Language Research (DeCLaRe) Lab
- neucodecNeuphonic
- neutts-2eNeuphonic
- neutts-airNeuphonic
- neutts-nanoNeuphonic
- omnivoice-singingModelsLab
- orpheus-3b-0.1-pretrainedCanopy Labs
- parler-tts-large-v1Parler TTS
- parler-tts-mini-expressoParler TTS
- parler-tts-mini-multilingual-v1.1Parler TTS
- parler-tts-mini-v1Parler TTS
- parler_tts_mini_v0.1Parler TTS
- personaplex-7b-v1NVIDIA
- personaplex-rl-seamlessKyutai
- phonon-31m-enGradium
- phonon-40m-enGradium
- phonon-40m-multilingualGradium
- pocket-tts-farsi-v2mehdi-hf
- realtime-tts-1.5-maxInworld AI$0.00005per character1 selling
- realtime-tts-1.5-miniInworld AI$0.000025per character1 selling
- resemble-enhanceResemble AI
- rumik-oss-1Rumik Intelligence Private Limited
- rvc-genshin-impactArkanDash
- s1-miniFish Audio$0.04→$0.08per Mtok in / out1 selling
- s2-proFish Audio$15per Mtok in2 selling
- sanoTTSAmpixa Labs
- sarashina2.2-ttsSB Intuitions
- sayro-tts-1.7BUzbek LLM Lab
- scenema-audioScenemaAI
- seamless-expressiveMeta
- seamless-streamingMeta
- sepformer-whamSpeechBrain
- sepformer-wsj02mixSpeechBrain
- silma-ttsSILMA AI
- sopro-v2-turbosamuel-vitorino
- speech-2.8-hdMiniMax9th of 41$0.0001per character1 selling
- speech-2.8-turboMiniMax14th of 41$0.00006per character2 selling
- speecht5_ttsMicrosoft
- speecht5_vcMicrosoft
- stable-audio-3-optimizedStability AI
- stable-audio-open-1.0Stability AI
- stable-audio-open-smallStability AI
- step-1o-audioStepFun$3.71→$8.9per Mtok in / out1 selling
- step-audio-2StepFun$1.48→$10.39per Mtok in / out1 selling
- step-audio-r1.5StepFun$1.48→$15.58per Mtok in / out1 selling
- step-tts-2StepFun$0.000042per character1 selling
- step-tts-miniStepFun$0.000013per character1 selling
- stepaudio-2.5-realtimeStepFun$1.48→$10.39per Mtok in / out1 selling
- stepaudio-2.5-ttsStepFun$0.000086per character1 selling
- supertonicSupertone
- supertonic-2Supertone
- supertonic-3Supertone
- tada-1bHume AI
- tada-3b-mlHume AI
- tangoDeep Cognition and Language Research (DeCLaRe) Lab
- ttsbalacoon$15per Mtok in1 selling
- tts-1.6b-en_frKyutai
- tts-tacotron2-ljspeechSpeechBrain
- tts_en_fastpitchNVIDIA
- tts_hifiganNVIDIA
- tts_transformer-zh-cv7_css10Meta
- vagdhenuprathoshap
- viXTTSNón Lá
- vits-ljsKakao Enterprise
- vuifluxions34th of 41
- xcodec2HKUST Audio