# Voicedex > Open catalog of speech AI: 815 speech-to-text, text-to-speech and speech-to-speech models and 15,499 voices from 126 providers, with supported languages, features, prices, regions and shutdown dates. Free, read-only JSON API; data is synced from provider APIs and docs and versioned in git. Prices are normalized for comparison: text-to-speech in USD per 1M characters, speech-to-text and speech-to-speech in USD per hour of audio. Non-USD prices are shown as published, never converted. Language codes are BCP-47; "*" means the provider supports many languages without listing them. ## Data - [Full catalog as JSON](https://voicedex.dev/v1/catalog.json): every provider, model and voice in one response - [API reference](https://voicedex.dev/docs): REST endpoints with filters for type, language, feature, region and price - [OpenAPI schema](https://voicedex.dev/openapi.json) - [MCP server](https://voicedex.dev/mcp): Streamable HTTP endpoint for AI assistants (search, compare, price models and voices) - [Source and data on GitHub](https://github.com/chandradot99/voicedex) ## Compare models - [Speech-to-text models](https://voicedex.dev/speech-to-text) - [Text-to-speech models](https://voicedex.dev/text-to-speech) - [Speech-to-speech models](https://voicedex.dev/speech-to-speech) - [All models, filterable](https://voicedex.dev/models) - [Voices with audio samples](https://voicedex.dev/voices) ## Providers - [2Noise (ChatTTS)](https://voicedex.dev/providers/2noise): 1 models - [Acapela Group](https://voicedex.dev/providers/acapela): 1 models, 124 voices - [AI for Thai (NECTEC)](https://voicedex.dev/providers/aiforthai): 2 models - [AI4Bharat](https://voicedex.dev/providers/ai4bharat): 3 models - [Aisha AI](https://voicedex.dev/providers/aisha): 2 models - [AISpeech (思必驰)](https://voicedex.dev/providers/aispeech): 8 models - [AITalk (AI Inc.)](https://voicedex.dev/providers/aitalk): 1 models - [Aivis Cloud API](https://voicedex.dev/providers/aivis): 1 models, 77 voices - [Alibaba Cloud Model Studio](https://voicedex.dev/providers/alibaba): 34 models, 164 voices - [Amazon Nova Sonic](https://voicedex.dev/providers/amazon-nova): 2 models, 16 voices - [Amazon Polly](https://voicedex.dev/providers/amazon-polly): 4 models, 119 voices - [Amazon Transcribe](https://voicedex.dev/providers/amazon-transcribe): 3 models - [AmiVoice](https://voicedex.dev/providers/amivoice): 17 models - [Aqua Voice (Avalon)](https://voicedex.dev/providers/aqua): 1 models - [AssemblyAI](https://voicedex.dev/providers/assemblyai): 7 models, 34 voices - [Async](https://voicedex.dev/providers/async): 4 models - [Azure Speech](https://voicedex.dev/providers/azure): 20 models, 1014 voices - [Baidu AI Cloud Speech](https://voicedex.dev/providers/baidu): 17 models - [Bhashini](https://voicedex.dev/providers/bhashini): 20 models - [Bland](https://voicedex.dev/providers/bland): 2 models, 3 voices - [Boson AI](https://voicedex.dev/providers/bosonai): 2 models, 6 voices - [Botnoi Voice](https://voicedex.dev/providers/botnoi): 3 models, 734 voices - [BreezeBlue](https://voicedex.dev/providers/breezeblue): 2 models - [BytePlus](https://voicedex.dev/providers/byteplus): 9 models, 326 voices - [CAMB.ai](https://voicedex.dev/providers/cambai): 7 models, 806 voices - [Canopy Labs](https://voicedex.dev/providers/canopylabs): 1 models - [Cartesia](https://voicedex.dev/providers/cartesia): 11 models, 913 voices - [CereProc](https://voicedex.dev/providers/cereproc): 1 models, 170 voices - [Chimege](https://voicedex.dev/providers/chimege): 4 models - [CoeFont](https://voicedex.dev/providers/coefont): 1 models, 34 voices - [Cohere](https://voicedex.dev/providers/cohere): 2 models - [Daglo (ActionPower)](https://voicedex.dev/providers/daglo): 1 models - [Deepdub](https://voicedex.dev/providers/deepdub): 3 models, 26 voices - [Deepgram](https://voicedex.dev/providers/deepgram): 44 models, 103 voices - [DeepL Voice API](https://voicedex.dev/providers/deepl): 1 models - [ElevenLabs](https://voicedex.dev/providers/elevenlabs): 18 models, 21 voices - [Fish Audio](https://voicedex.dev/providers/fishaudio): 8 models, 664 voices - [FPT.AI](https://voicedex.dev/providers/fptai): 2 models, 7 voices - [FunAudioLLM (Alibaba Tongyi)](https://voicedex.dev/providers/funaudiollm): 3 models - [Gladia](https://voicedex.dev/providers/gladia): 2 models - [Gnani.ai](https://voicedex.dev/providers/gnani): 5 models, 76 voices - [Google Cloud](https://voicedex.dev/providers/google-cloud): 41 models, 2066 voices - [Gradium](https://voicedex.dev/providers/gradium): 4 models, 1432 voices - [Groq](https://voicedex.dev/providers/groq): 4 models, 12 voices - [Hexgrad](https://voicedex.dev/providers/hexgrad): 2 models - [Huawei Cloud SIS](https://voicedex.dev/providers/huawei-cloud): 6 models, 4 voices - [Humelo](https://voicedex.dev/providers/humelo): 2 models - [iApp Technology](https://voicedex.dev/providers/iapp): 7 models, 3 voices - [IBM Watson](https://voicedex.dev/providers/ibm): 54 models, 50 voices - [iFLYTEK](https://voicedex.dev/providers/iflytek): 3 models, 61 voices - [IndexTTS (bilibili)](https://voicedex.dev/providers/index-tts): 2 models - [Intron Health](https://voicedex.dev/providers/intron): 2 models - [Inworld AI](https://voicedex.dev/providers/inworld): 11 models, 366 voices - [k2-fsa (Next-gen Kaldi)](https://voicedex.dev/providers/k2-fsa): 1 models - [Khaya AI (GhanaNLP)](https://voicedex.dev/providers/khaya): 2 models, 3 voices - [KittenML](https://voicedex.dev/providers/kittenml): 8 models - [Krutrim](https://voicedex.dev/providers/krutrim): 3 models - [KugelAudio](https://voicedex.dev/providers/kugelaudio): 5 models, 169 voices - [Kyutai](https://voicedex.dev/providers/kyutai): 1 models - [Lelapa AI](https://voicedex.dev/providers/lelapa): 2 models - [Liquid AI](https://voicedex.dev/providers/liquidai): 3 models - [Maya Research](https://voicedex.dev/providers/maya-research): 6 models - [Meta](https://voicedex.dev/providers/meta): 1 models - [Microsoft (open weights)](https://voicedex.dev/providers/microsoft): 7 models - [MiniMax](https://voicedex.dev/providers/minimax): 8 models, 332 voices - [Mistral AI](https://voicedex.dev/providers/mistral): 7 models - [Modulate](https://voicedex.dev/providers/modulate): 3 models - [Munsit](https://voicedex.dev/providers/munsit): 3 models, 176 voices - [Murf AI](https://voicedex.dev/providers/murf): 2 models, 162 voices - [MyShell](https://voicedex.dev/providers/myshell-ai): 5 models - [Nari Labs](https://voicedex.dev/providers/nari-labs): 2 models - [Navana.ai](https://voicedex.dev/providers/navana): 12 models, 55 voices - [Naver CLOVA](https://voicedex.dev/providers/naver-clova): 5 models, 108 voices - [Neuphonic](https://voicedex.dev/providers/neuphonic): 1 models, 120 voices - [NVIDIA](https://voicedex.dev/providers/nvidia): 15 models, 101 voices - [OpenAI](https://voicedex.dev/providers/openai): 29 models, 34 voices - [OpenBMB](https://voicedex.dev/providers/openbmb): 3 models - [OpenMOSS (Fudan)](https://voicedex.dev/providers/openmoss): 5 models - [Oracle OCI Speech](https://voicedex.dev/providers/oracle-oci): 6 models - [Parler-TTS (Hugging Face)](https://voicedex.dev/providers/parler-tts): 3 models - [Phonexia](https://voicedex.dev/providers/phonexia): 2 models - [Phonic](https://voicedex.dev/providers/phonic): 1 models, 39 voices - [Picovoice](https://voicedex.dev/providers/picovoice): 3 models, 15 voices - [Proto](https://voicedex.dev/providers/proto): 2 models - [ReadSpeaker](https://voicedex.dev/providers/readspeaker): 1 models, 282 voices - [Reson8](https://voicedex.dev/providers/reson8): 3 models - [Respeecher](https://voicedex.dev/providers/respeecher): 8 models, 131 voices - [ReturnZero (RTZR)](https://voicedex.dev/providers/rtzr): 4 models - [Rev AI](https://voicedex.dev/providers/revai): 1 models - [Reverie](https://voicedex.dev/providers/reverie): 4 models, 41 voices - [Rime AI](https://voicedex.dev/providers/rime): 4 models, 444 voices - [RVC-Boss (GPT-SoVITS)](https://voicedex.dev/providers/rvc-boss): 1 models - [Sarj.ai](https://voicedex.dev/providers/sarj): 2 models - [Sarvam](https://voicedex.dev/providers/sarvam): 7 models, 266 voices - [Sesame](https://voicedex.dev/providers/sesame): 1 models - [Sestek (Knovvu)](https://voicedex.dev/providers/sestek): 47 models, 99 voices - [Shunya Labs](https://voicedex.dev/providers/shunya): 7 models, 46 voices - [SK Telecom (A.X)](https://voicedex.dev/providers/skt): 4 models - [Smallest AI](https://voicedex.dev/providers/smallestai): 6 models, 508 voices - [Soniox](https://voicedex.dev/providers/soniox): 5 models, 200 voices - [Speechify](https://voicedex.dev/providers/speechify): 5 models, 1108 voices - [Speechmatics](https://voicedex.dev/providers/speechmatics): 5 models - [Spitch](https://voicedex.dev/providers/spitch): 2 models, 28 voices - [StepFun](https://voicedex.dev/providers/stepfun): 7 models, 7 voices - [Sunbird AI](https://voicedex.dev/providers/sunbird): 2 models, 48 voices - [Suno (Bark)](https://voicedex.dev/providers/suno): 2 models - [SWivid (F5-TTS)](https://voicedex.dev/providers/swivid): 2 models - [T-Bank VoiceKit](https://voicedex.dev/providers/tbank): 3 models - [Tencent Cloud](https://voicedex.dev/providers/tencent): 18 models, 56 voices - [Typecast AI](https://voicedex.dev/providers/typecastai): 3 models, 636 voices - [Unisound (云知声)](https://voicedex.dev/providers/unisound): 4 models - [Unreal Speech](https://voicedex.dev/providers/unrealspeech): 1 models, 52 voices - [Uplift AI](https://voicedex.dev/providers/upliftai): 3 models, 91 voices - [Vatis Tech](https://voicedex.dev/providers/vatis): 1 models - [Vbee](https://voicedex.dev/providers/vbee): 1 models - [Voice.ai](https://voicedex.dev/providers/voiceai): 6 models - [Voicegain](https://voicedex.dev/providers/voicegain): 9 models - [Volcengine](https://voicedex.dev/providers/volcengine): 15 models, 601 voices - [Voxist](https://voicedex.dev/providers/voxist): 6 models - [WellSaid Labs](https://voicedex.dev/providers/wellsaid): 2 models - [xAI](https://voicedex.dev/providers/xai): 4 models, 28 voices - [Xiaomi MiMo](https://voicedex.dev/providers/xiaomi): 4 models, 9 voices - [Yandex SpeechKit](https://voicedex.dev/providers/yandex): 9 models, 34 voices - [Zhipu AI (Z.ai)](https://voicedex.dev/providers/zhipu): 5 models, 12 voices - [Zoom AI Services](https://voicedex.dev/providers/zoom): 4 models - [Zyphra](https://voicedex.dev/providers/zyphra): 2 models, 27 voices ## Optional Providers that shut down or stopped selling (kept for the record): - [Coqui](https://voicedex.dev/providers/coqui) - [Dubverse](https://voicedex.dev/providers/dubverse) - [Hume AI](https://voicedex.dev/providers/humeai) - [LMNT](https://voicedex.dev/providers/lmnt) - [Lovo AI](https://voicedex.dev/providers/lovoai) - [Nijivoice](https://voicedex.dev/providers/nijivoice) - [PlayHT](https://voicedex.dev/providers/playht) - [Replica Studios](https://voicedex.dev/providers/replica) - [Resemble AI](https://voicedex.dev/providers/resemble) - [Sber SaluteSpeech](https://voicedex.dev/providers/sber-salutespeech) - [Supertone](https://voicedex.dev/providers/supertone) - [Symbl.ai](https://voicedex.dev/providers/symbl) - [Verbio](https://voicedex.dev/providers/verbio)