Chariot-1-ASR Realtime
Voice typing in Hindi, and 11 more languages.
Speak into your microphone and the words appear while you are still talking. Copy the text into a message, a document or a form. Hindi with English words, Tamil, Bengali and nine more languages.
Free to start · No card · In your browser · 12 languages
The clip
English words
Clip: FLEURS, CC BY 4.0
#1in Hindi, against every streaming API and every batch model we tested
23%fewer errors in Hindi than Sarvam Saaras v4, the next best streaming API
12languages, with English words in Latin letters or in the language’s own script
10,000free credits to start, almost two hours of speech
How it works
Talk, then copy.
Open Speech to Text
Sign up free and open Speech to Text in the web app, in your browser.
Press record and speak
Allow the microphone and talk. The words appear as you say them, in Hindi or any of 12 languages.
Copy the text
Copy the transcript and paste it wherever you need it. Every session is saved in your history.
API
From audio to text in a few lines.
One WebSocket. Send audio as you capture it, read text as it is recognised, and send input.done for the complete transcript.
# pip install websockets
import json, websockets
url = ("wss://api.chariot.in/v1/asr/ws"
"?language=hi&audio_format=pcm_16000&mode=high_accuracy")
headers = {"chariotai-api-key": API_KEY}
async with websockets.connect(url, additional_headers=headers) as ws:
await ws.recv() # session.started
for chunk in microphone(): # 100 ms of audio at a time
await ws.send(chunk)
await ws.send(json.dumps({"type": "input.done"}))
async for message in ws:
event = json.loads(message)
if event["type"] == "transcript.delta":
print(event["text"], end="")Dictation
Write by speaking.
Hindi with English words
Codemix keeps English words in Latin letters; native writes every word in Devanagari.
High accuracy for dictation
High accuracy takes more context before it commits text, for dictation where accuracy matters more than a short delay.
Keep talking
An option finalises each phrase after a pause and keeps listening until you stop recording.
Any of 12 languages
Set your language, or leave it out and the model detects it from the audio.
In your own app
Build dictation into your product with the same model, over one WebSocket.
Processed in India
Audio and transcripts are processed on Indian infrastructure, under the Digital Personal Data Protection Act.
Pricing
Plans from ₹499 a month.Plans from $5 a month. Or start free.
Speech to text uses the same credits as text to speech, at one rate in both modes. Each plan covers this much audio a month.
| Plan | A month | Speech to text |
|---|---|---|
| Free10,000 credits | ₹0$0 | about 2 hours |
| Starter100,000 credits | ₹499$5 | about 19 hours |
| Creator250,000 credits | ₹999 | about 46 hours |
| Startup1,500,000 credits | ₹4,999$49 | about 280 hours |
| Scale10,000,000 credits | ₹29,999$299 | about 1,900 hours |
Monthly prices. Prices exclude applicable taxes.
Use cases
Other things people build with it.
Voice agents, too: they hear a caller while they are still speaking, in their language.
Audio and video to text
Upload a recording or a video and get its transcript, in the language it was spoken.
Audio and video to text ›Call transcription
Support and sales calls written down, live from the phone line or from recordings.
Call transcription ›Your data
Your audio stays in India.
Speech to text runs on Indian infrastructure. Audio you send and its transcripts are processed in India under the Digital Personal Data Protection Act. Chariot is one of four teams selected under the IndiaAI Mission to build the country’s foundation models.
Is it free?
Every account starts with 10,000 free credits, almost two hours of speech, with no card. After that, speech to text costs 1.5 credits a second.
Does it replace my phone keyboard?
No. It works in the Chariot web app in your browser, and in your own app through the API. Copy the text from there.
Can I type Hindi with English words?
Yes. Codemix, the default, keeps English words in Latin letters; native writes them in Devanagari. Choose before you start.
Which languages can I speak?
Hindi, Marathi, Bengali, Tamil, Telugu, Kannada, Malayalam, Gujarati, Punjabi, Odia, Sanskrit and English.
What is the difference between balanced and high accuracy?
Balanced sends text sooner, for voice agents and live captions. High accuracy takes more context before it commits text, which usually gives a more accurate transcript, for calls, meetings and dictation. Both cost the same.
Where is my audio processed?
In India. Audio you send and its transcripts are processed on Indian infrastructure, under the Digital Personal Data Protection Act.
