Text to speech
curl --request POST \
--url https://api.chariot.in/v1/tts \
--header 'Content-Type: application/json' \
--header 'chariotai-api-key: <api-key>' \
--data '
{
"voice_id": "bac7d666-094d-4698-91fa-741d60fce662",
"text": "Hello from Chariot.",
"model_type": "chariot-multilingual-v1",
"speed": 1,
"optimize_streaming_latency": 0
}
'import requests
url = "https://api.chariot.in/v1/tts"
payload = {
"voice_id": "bac7d666-094d-4698-91fa-741d60fce662",
"text": "Hello from Chariot.",
"model_type": "chariot-multilingual-v1",
"speed": 1,
"optimize_streaming_latency": 0
}
headers = {
"chariotai-api-key": "<api-key>",
"Content-Type": "application/json"
}
response = requests.post(url, json=payload, headers=headers)
print(response.text)const options = {
method: 'POST',
headers: {'chariotai-api-key': '<api-key>', 'Content-Type': 'application/json'},
body: JSON.stringify({
voice_id: 'bac7d666-094d-4698-91fa-741d60fce662',
text: 'Hello from Chariot.',
model_type: 'chariot-multilingual-v1',
speed: 1,
optimize_streaming_latency: 0
})
};
fetch('https://api.chariot.in/v1/tts', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));<?php
$curl = curl_init();
curl_setopt_array($curl, [
CURLOPT_URL => "https://api.chariot.in/v1/tts",
CURLOPT_RETURNTRANSFER => true,
CURLOPT_ENCODING => "",
CURLOPT_MAXREDIRS => 10,
CURLOPT_TIMEOUT => 30,
CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
CURLOPT_CUSTOMREQUEST => "POST",
CURLOPT_POSTFIELDS => json_encode([
'voice_id' => 'bac7d666-094d-4698-91fa-741d60fce662',
'text' => 'Hello from Chariot.',
'model_type' => 'chariot-multilingual-v1',
'speed' => 1,
'optimize_streaming_latency' => 0
]),
CURLOPT_HTTPHEADER => [
"Content-Type: application/json",
"chariotai-api-key: <api-key>"
],
]);
$response = curl_exec($curl);
$err = curl_error($curl);
curl_close($curl);
if ($err) {
echo "cURL Error #:" . $err;
} else {
echo $response;
}package main
import (
"fmt"
"strings"
"net/http"
"io"
)
func main() {
url := "https://api.chariot.in/v1/tts"
payload := strings.NewReader("{\n \"voice_id\": \"bac7d666-094d-4698-91fa-741d60fce662\",\n \"text\": \"Hello from Chariot.\",\n \"model_type\": \"chariot-multilingual-v1\",\n \"speed\": 1,\n \"optimize_streaming_latency\": 0\n}")
req, _ := http.NewRequest("POST", url, payload)
req.Header.Add("chariotai-api-key", "<api-key>")
req.Header.Add("Content-Type", "application/json")
res, _ := http.DefaultClient.Do(req)
defer res.Body.Close()
body, _ := io.ReadAll(res.Body)
fmt.Println(string(body))
}HttpResponse<String> response = Unirest.post("https://api.chariot.in/v1/tts")
.header("chariotai-api-key", "<api-key>")
.header("Content-Type", "application/json")
.body("{\n \"voice_id\": \"bac7d666-094d-4698-91fa-741d60fce662\",\n \"text\": \"Hello from Chariot.\",\n \"model_type\": \"chariot-multilingual-v1\",\n \"speed\": 1,\n \"optimize_streaming_latency\": 0\n}")
.asString();require 'uri'
require 'net/http'
url = URI("https://api.chariot.in/v1/tts")
http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true
request = Net::HTTP::Post.new(url)
request["chariotai-api-key"] = '<api-key>'
request["Content-Type"] = 'application/json'
request.body = "{\n \"voice_id\": \"bac7d666-094d-4698-91fa-741d60fce662\",\n \"text\": \"Hello from Chariot.\",\n \"model_type\": \"chariot-multilingual-v1\",\n \"speed\": 1,\n \"optimize_streaming_latency\": 0\n}"
response = http.request(request)
puts response.read_body"<string>"Text to speech
Text to speech
Synthesize speech from text and receive the complete audio file in the response body as a single audio/wav payload. Best for short, non-latency-critical text (up to 700 characters for chariot-v0, or 1,000 for chariot-multilingual-v1). For first-byte latency, use POST /v1/tts/stream or the WebSocket API instead.
The response body is the raw WAV audio. Metadata is returned in response headers: Inference-Id, Audio-Url (a persistent link to the same audio), and Credit-Utilized (credits charged for this request).
Cost: 1 credit per input character.
Pass model_type in the JSON body. It defaults to chariot-v0. Sync splits input internally when needed and returns one assembled WAV. Unsupported models, voice/model combinations, or speeds return 422.
POST
/
v1
/
tts
Text to speech
curl --request POST \
--url https://api.chariot.in/v1/tts \
--header 'Content-Type: application/json' \
--header 'chariotai-api-key: <api-key>' \
--data '
{
"voice_id": "bac7d666-094d-4698-91fa-741d60fce662",
"text": "Hello from Chariot.",
"model_type": "chariot-multilingual-v1",
"speed": 1,
"optimize_streaming_latency": 0
}
'import requests
url = "https://api.chariot.in/v1/tts"
payload = {
"voice_id": "bac7d666-094d-4698-91fa-741d60fce662",
"text": "Hello from Chariot.",
"model_type": "chariot-multilingual-v1",
"speed": 1,
"optimize_streaming_latency": 0
}
headers = {
"chariotai-api-key": "<api-key>",
"Content-Type": "application/json"
}
response = requests.post(url, json=payload, headers=headers)
print(response.text)const options = {
method: 'POST',
headers: {'chariotai-api-key': '<api-key>', 'Content-Type': 'application/json'},
body: JSON.stringify({
voice_id: 'bac7d666-094d-4698-91fa-741d60fce662',
text: 'Hello from Chariot.',
model_type: 'chariot-multilingual-v1',
speed: 1,
optimize_streaming_latency: 0
})
};
fetch('https://api.chariot.in/v1/tts', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));<?php
$curl = curl_init();
curl_setopt_array($curl, [
CURLOPT_URL => "https://api.chariot.in/v1/tts",
CURLOPT_RETURNTRANSFER => true,
CURLOPT_ENCODING => "",
CURLOPT_MAXREDIRS => 10,
CURLOPT_TIMEOUT => 30,
CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
CURLOPT_CUSTOMREQUEST => "POST",
CURLOPT_POSTFIELDS => json_encode([
'voice_id' => 'bac7d666-094d-4698-91fa-741d60fce662',
'text' => 'Hello from Chariot.',
'model_type' => 'chariot-multilingual-v1',
'speed' => 1,
'optimize_streaming_latency' => 0
]),
CURLOPT_HTTPHEADER => [
"Content-Type: application/json",
"chariotai-api-key: <api-key>"
],
]);
$response = curl_exec($curl);
$err = curl_error($curl);
curl_close($curl);
if ($err) {
echo "cURL Error #:" . $err;
} else {
echo $response;
}package main
import (
"fmt"
"strings"
"net/http"
"io"
)
func main() {
url := "https://api.chariot.in/v1/tts"
payload := strings.NewReader("{\n \"voice_id\": \"bac7d666-094d-4698-91fa-741d60fce662\",\n \"text\": \"Hello from Chariot.\",\n \"model_type\": \"chariot-multilingual-v1\",\n \"speed\": 1,\n \"optimize_streaming_latency\": 0\n}")
req, _ := http.NewRequest("POST", url, payload)
req.Header.Add("chariotai-api-key", "<api-key>")
req.Header.Add("Content-Type", "application/json")
res, _ := http.DefaultClient.Do(req)
defer res.Body.Close()
body, _ := io.ReadAll(res.Body)
fmt.Println(string(body))
}HttpResponse<String> response = Unirest.post("https://api.chariot.in/v1/tts")
.header("chariotai-api-key", "<api-key>")
.header("Content-Type", "application/json")
.body("{\n \"voice_id\": \"bac7d666-094d-4698-91fa-741d60fce662\",\n \"text\": \"Hello from Chariot.\",\n \"model_type\": \"chariot-multilingual-v1\",\n \"speed\": 1,\n \"optimize_streaming_latency\": 0\n}")
.asString();require 'uri'
require 'net/http'
url = URI("https://api.chariot.in/v1/tts")
http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true
request = Net::HTTP::Post.new(url)
request["chariotai-api-key"] = '<api-key>'
request["Content-Type"] = 'application/json'
request.body = "{\n \"voice_id\": \"bac7d666-094d-4698-91fa-741d60fce662\",\n \"text\": \"Hello from Chariot.\",\n \"model_type\": \"chariot-multilingual-v1\",\n \"speed\": 1,\n \"optimize_streaming_latency\": 0\n}"
response = http.request(request)
puts response.read_body"<string>"Authorizations
Your Chariot API key, generated from the dashboard. Send it in the chariotai-api-key header on every request.
Body
application/json
The voice to synthesize with. Get valid ids from GET /v1/voices.
Example:
"bac7d666-094d-4698-91fa-741d60fce662"
Text to synthesize, with at least one character after trimming surrounding whitespace. For chariot-v0, sync accepts up to 700 characters and HTTP streaming up to 500. For chariot-multilingual-v1, both accept up to 1,000.
Required string length:
1 - 1000Example:
"Hello from Chariot."
TTS model to use. Defaults to chariot-v0. The legacy spelling v0 remains accepted for compatibility. Select a model supported by the voice.
Available options:
chariot-v0, chariot-multilingual-v1 Speaking rate. Choose a value from the selected voice model entry's speeds list. chariot-v0 supports only 1.0. Other rates depend on the voice and model; an unavailable rate returns 422.
Required range:
0.5 <= x <= 2Latency-optimization level. 0 applies all quality passes. Higher values trade quality passes for lower latency and may reduce output quality.
Required range:
0 <= x <= 4Response
The synthesized speech as a WAV file.
The response is of type file.