API kupu ki te kōrero mō ngā kaiwhakawhanake

Ka hanga i ngā taupānga kōrero ā-waha me tātau REST API. Ka tāpiri i te kupu-ki-te-kōrero māori, te tāruatanga reo, te kōrero-ki-te-tuhi, me te tukanga oro ki ōna taupānga, ngā tāngata kōrero, ngā kaiāwhina reo, me ngā hua SaaS. He āhua ōrite OpenAI, 24+ ngā tauira, he whakakotahitanga ngāwari.

REST API Chatbots Whakataupānga reo Huahana SaaS Mā te whakamātautau

Ka whakamātautia ināianei

0/500
Waihoki me Kokoro, Piper, VITS, MeloTTS
Your generated audio will appear here
I hangaia
0:00 0:00
Waihoki
Pērā ki a TTS.ai? E kōrero ana ki ōna hoa!

Āhuahira API mō ngā kaiwhakawhanake

Ko ngā mea katoa e hiahiatia ana e koe hei hanga i ngā taupānga kōrero āhei

He ngāwari te API REST

Ko te tono POST kotahi hei waihanga i te kōrero. Te tono JSON, te urupare oro. Ka mahi ki tētahi reo papatono e tautoko ana i te HTTP.

OpenAI-Hoatu

Ka whakahōu-i te whakahōutanga mō OpenAI TTS API. Ka whakawhiti i tōna pūtake_url me te kī API — ka mahi tere te waehere tīariari.

24+ ngā tauira e wātea ana

Ka uru ki ia tauira mā tētahi API kotahi. Ka huri i ngā tauira mā te huri i tētahi tohuāhua. E whakataurite ana i te āhuatanga, te tere, me te utu.

Sub-Second Latency

Ko te Kokoro e waihanga ai te oro i raro iho i te 1 sekone. He tino pai mō ngā tāngata kōrero, ngā kaiāwhina reo, me ngā taupānga whakawhitiwhitinga.

Te API Cloning reo

Ka tārua ētahi oro mai i tētahi tauira oro poto mā te API. Ka whakamahia ngā oro tārua mō ngā whakatupuranga o muri ake katoa.

He maha nga hanga

Whakaputa hei WAV, MP3, OGG, FLAC rānei. Hiko i te mokatere tauira me te hōhonu iti. Whakawhiwhinga tautoko mō ngā taupānga wā tūturu.

Ko ngā tauira pai rawa mō te whakakotahitanga kaiwhakawhanake

Ka kōwhiria te tauira tika mō tōna taupānga

KokoroKokoro

Free

Lightweight 82M parameter model delivering studio-quality speech with blazing-fast inference.

Fast 5/5

Ko te tino pai mo: Tauira tere rawa - sub-second latency, tino pai mo ngā taupānga wā tūturu me ngā tāngata kōrero

Whakamātautau Kokoro

CosyVoice 2CosyVoice 2

Standard

Alibaba's scalable streaming TTS with human-parity naturalness and near-zero latency.

Medium 5/5 Ko te tāruatanga reo

Ko te tino pai mo: TTS Streaming me te tārua reo mō ngā taupānga āwhina reo

Whakamātautau CosyVoice 2

Sesame CSMSesame CSM

Premium

Conversational speech model generating natural dialogue with appropriate timing and emotion.

Slow 5/5

Ko te tino pai mo: AI kōrerorero me te wā māori mō te kōrerorero me te reo āwhina

Whakamātautau Sesame CSM

PiperPiper

Free

A fast, local neural text to speech system optimized for Raspberry Pi and embedded devices.

Fast 3/5

Ko te tino pai mo: Waihoki, tauira CPU-only mō ngā taupānga nui rawa me te utu pūtea kore

Whakamātautau Piper

BarkBark

Standard

Transformer-based text-to-audio model that generates realistic speech, music, and sound effects.

Slow 4/5

Ko te tino pai mo: Ko te whakatūnga oro me ngā pānga oro mō ngā taupānga auaha me te whakangahau

Whakamātautau Bark

He pēhea te whakauru i te TTS API

Mai i te whakaingoatanga ki te huaina API tuatahi i raro i te 5 min

1

Ki te whiwhi i ōna kī API

Ka tāuru mō te wāteatanga me te waihanga i tētahi kī API mai i tōmu papatono kāwai. 50 ngā pūtea i whakaurua.

2

Whakahohe i Ōtou Hiranga Tuatahi

POST ki te /v1/tts me te kupu, te tauira, me te reo. Ka whiwhi taipitopito oro. I raro i te 5 rārangi waehere.

3

Hiku i tōna tauira

Mā te whakamātau i ngā tauira rerekē mō tōmu take whakamahi. E whakataurite ana i te tere, te āhuatanga, me te utu i ia whakatupuranga.

4

Ka tukuna ki te whakanaotanga

Mā te utu-i-te-kau. Kāore he tepe ōrautanga i runga i ngā mahere utu. Whakamātautau i te whakamahinga i roto i tōmu papatono.

Ko ngā tauira waehere tīmata tere

Kohi TTS.ai i roto i tētahi reo me tātau REST API

Python I rongonuitia
import requests

response = requests.post(
    "https://api.tts.ai/v1/tts",
    json={
        "text": "Hello from my app!",
        "model": "kokoro",
        "voice": "af_heart",
        "format": "mp3"
    },
    headers={
        "Authorization": "Bearer sk-tts-xxx"
    }
)

with open("output.mp3", "wb") as f:
    f.write(response.content)
JavaScript (Node.js) Node.js
const response = await fetch(
    "https://api.tts.ai/v1/tts",
    {
        method: "POST",
        headers: {
            "Content-Type": "application/json",
            "Authorization": "Bearer sk-tts-xxx"
        },
        body: JSON.stringify({
            text: "Hello from my app!",
            model: "kokoro",
            voice: "af_heart",
            format: "mp3"
        })
    }
);

const audio = await response.blob();
cURL Ko te katoa
curl -X POST https://api.tts.ai/v1/tts \
  -H "Authorization: Bearer sk-tts-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Hello from my app!",
    "model": "kokoro",
    "voice": "af_heart",
    "format": "mp3"
  }' \
  --output output.mp3
OpenAI-Hoatu te hanga e ōrite ana Ka tūpāpaku
# Works with OpenAI client library
from openai import OpenAI

client = OpenAI(
    api_key="sk-tts-xxx",
    base_url="https://api.tts.ai/v1"
)

response = client.audio.speech.create(
    model="kokoro",
    voice="af_heart",
    input="Hello from my app!"
)

response.stream_to_file("output.mp3")

He aha nga kaiwhakawhanake e hanga ana me TTS.ai

Ko ngā tauira whakakotahitanga noa me ngā taupānga

AI Chatbots me ngā Kaiāwhina

E tāpiri ana i te huaputa reo ki tōna kaitiaki AI rānei. Ko ngā urupare LLM i roto i te TTS mō ngā whakawhitinga reo-ka taea. Ko te Kokoro e tuku ana i te tūponotanga o te tuarua mō ngā kōrero i te wā tūturu. Ko te Sesame CSM e whakaputa ana i te kōrero me te wā māori.

  • Ka urupare a LLM ki te whakawhitinga kōrero
  • Ko te ātetetanga o te tuarua-roto me Kokoro
  • Ka kōrerorerotia te kōrero me te Sesame CSM
  • Ko te huaputa oro

Pāpāho me ngā taupānga reo

Ka hanga i ngā taupānga pūkoro ā-waha, ngā utauta āhei, ngā taupānga whakaakoranga, me ngā pāpāho akoranga reo. Ka mahi a tātau REST API me tētahi anga pāpāho. Whakataki i ngā faila oro, te rerenga rānei ki te kaiwhakahaere.

  • React Native, Flutter, Swift, Kotlin
  • Kei te āheitanga me ngā taupānga akoranga
  • Pānga akoranga reo
  • Hanganga ihirangi oro

Huahana SaaS

Mā ngā āheinga reo tohu-mārō i roto i kāu hua SaaS. Tāpiri i te TTS, STT, te tārua reo, me te tukanga oro hei āhuahira i roto i kāu papatono. Ka whakamahia e tātau te API hei ārai reo kāore i te whakahaere i ngā hanganga GPU.

  • Mā ngā āhuatanga reo tohu-mā
  • Kāore he hanganga GPU e hiahiatia ana
  • Ko te utu-i-te-whakahaeretanga
  • 24+ ngā tauira hei whakarato i ōna kaimahi

Automation Pipelines

Ka whakaurua te whakanao reo ki roto i ngā pūwhitinga CI/CD, te whakamātautau ihirangi, me ngā rerenga mahi whakamātautau. Whakanao i ngā mano o ngā pūkete oro mai i ngā raraunga papatono, te whakanao i te whakanao podcast, te hanga rānei i ngā pūwhitinga tauwāhi ihirangi.

  • Ka whakatinana te rōpū mā te API
  • Ko ngā pūwhitinga tauwāhi ihirangi
  • Ko te whakakotahitanga CI/CD
  • Te ripanga ki te whakamātautau reo

Ko ngā whakaritenga API

I hangaia mō ngā taupānga whakaputa

24+

Kāhua TTS

100+

Pāpāho

30+

reo

<1s

Te ātetetanga (Kokoro)

E pā ana ngā pātai

Ko ngā pātai noa iho mo te TTS.ai developer API

He. Ko tātau API e whai ana i te āhua kōrero reo OpenAI. Mēnā e whakamahi ana koe i te OpenAI Python, i te puna kākāriki JavaScript rānei, ka taea e koe te whakawhiti ki te TTS.ai mā te huri i te pūtake_url me ngā tohu api_key. Ka mahi tōtou waehere o nāianei me te kore whakarerekētanga.

Ko te Kokoro e whakanao ana i te oro i raro i te 1 sekone mō ngā rerenga pūnoa. Ko te CosyVoice 2 e tautoko ana i te huaputa rerenga mō te ātetetanga iti iho. Mō ngā tāngata kōrero me ngā kaiāwhina reo, ko te nuinga o te wā rererangi ko te 1–3 sekone, i runga anō i te roanga kupu me te kōwhiringa tauira.

Ko ngā tauira wātea (Kokoro, Piper, VITS, MeloTTS) e utu ana i ngā pūtea kore. Ko ngā tauira paerewa e utu ana i ngā pūtea 2 mō ia pūāhua 1,000. Ko ngā tauira utu e utu ana i ngā pūtea 4 mō ia pūāhua 1,000. Ko ngā pūnaha e tīmata ana i te $9 / mahina mō ngā pūtea 500.

Ināianei, tuku i tētahi tauira oro tohutoro (5-30 sekone) ki te wāhi mutunga o te tārua reo, kātahi ka whakamahi i te ID reo tārua i roto i ngā tono TTS o muri ake. Ko ngā tauira e tautoko ana i te tārua ko te CosyVoice 2, Chatterbox, Fish Speech, me GPT-SoVITS.

Ko te taumata wātea wātea e whakawhāiti ana i te mokatere taketake (3 ngā tono i ia wā me te kore pūkete). He whakawhāiti mokatere nui ngā mahere utu e hāngai ana ki ngā taupānga whakanaotanga. Tātai mai ki a mātou mō ngā whakaritenga tika o te taumata umanga.

WAV (kāore i te whakawhāititia, te āhuatanga tiketike rawa), MP3 (whakawhāititia, ngā pūranga iti iho), OGG (whakawhāititia), me te FLAC (whakawhāititia kāore i te ngaro). Ka whakapūtātia te hanga i roto i tātou tono. Ko te pūtake ko te WAV i te mokatere tauira taketake o te tauira.

Heoi. Ko te whakakotahitanga o tātau TTS API me tētahi tauira kōrero-ki-tuhi me tētahi LLM hei hanga i tētahi pūwhitinga āwhina reo katoa. Ko te Kokoro e whakarato ana i te tūponotanga o te tuarua mō te kōrero i te wā tūturu. Ko te CosyVoice 2 e tautoko ana i te huaputa rerenga mō ngā wā urupare iti iho.

Ko te CosyVoice 2 me te Kokoro e tautoko ana i te whakaputanga oro whakatere i reira ka tukua nga pouaka oro i a rātau e whakaputaina ana. Ka whakaiti tēnei i te wā-ki-te-pae tuatahi mō ngā taupānga wā tūturu pēnei i ngā kaiāwhina reo me ngā wheako whakawhitiwhiti.

E hoki ana te API ki ngā waehere tūnga HTTP paerewa. Ka whakatinana te whakateretanga mō ngā hapa 5xx me ngā urupare tepe. Mō ngā taupānga hira, tāpiri i tētahi raupapa me ngā arorau whakamātau anō. He nui te API o tātau, engari ka whakatūpatotia i ngā wā katoa te whakahaere hapa.

Ināianei. Ko ngā /v1/voices me ngā /v1/models e hoki ana ki ngā rārangi JSON o ngā reo me ngā tauira katoa e wātea ana me ā rātou metadata (whakaahua reo, whakarārangi tika, whakarārangi tere, me te taumata utu). Ka whakamahia ēnei hei hanga i ngā kōwhiringa tauira hihiri i roto i kāu taupānga.

Ko ngā tauira wāteatanga (Kokoro, Piper, VITS, MeloTTS) e mahi ana hei kāwai mārama mai i a rātau e utu ana i ngā pūtea kore. Whakamātau i tō tātou whakauru ki ngā tauira wāteatanga, kātahi ka huri ki ngā tauira utu i roto i te whakanaotanga mā te huri i te tātai tauira. Kāore he taiao whakamātautau motuhake e hiahiatia ana.

Ko te nuinga o a tātau tauira he pūtake tūwhera, ā, ka taea te whakawhiwhia ki a ia anō. Heoi anō, e hiahiatia ana e te whakawhiwhinga ki a ia anō ngā rauemi GPU nui (ka whakamahia e tātau te 4x NVIDIA Tesla P40 me te 96GB VRAM katoa).
5.0/5 (1)

E whakaritea ana ki te hanga me te AI reo?

Ki te whiwhi i tātou kī API wātea me te tīmata i te hanganga. 50 ngā pūtea i te whakaingoatanga, ngā tauira wātea, ngā tuhinga whānui.