Chariot-1-ASR Realtime
Sanskrit speech to text in real time.
Spoken Sanskrit, written down as it is spoken: live from a microphone, or from a recording. Sanskrit is one of the 12 languages in Chariot-1-ASR Realtime, at the same price as the others.
Free to start · No card · Live audio or file upload · language=sa
12languages in one model, Sanskrit among them
saits language code: set it, or leave it out and the model detects the language
1.5credits a second of audio, the same as every language
3 ha streaming session, or a recording through the async API
Live audio and recordings
Set it once per session. Then talk.
Each session takes a mode, a language or none, and how English words are written. One price for all of it.
Balanced
Text arrives sooner. For voice agents, live captions and anything a person reacts to as they speak.
High accuracy
The model takes more context before it commits text. For call and meeting transcripts and dictation, where accuracy matters more than a short delay.
Any language, or the one you set
Set the language when you know what callers speak, or leave it out and the model detects it from the audio.
Codemix or native
Codemix keeps borrowed English words in Latin script. Native writes them in the script of the spoken language.
Phone audio
8 kHz mu-law and A-law straight from a phone line, or 16-bit PCM from 8 to 48 kHz.
Recordings too
Upload a recording in the web app and get its transcript back. WAV, MP3, FLAC, OGG, M4A, AAC or WebM.
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=sa&audio_format=pcm_16000&mode=balanced")
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="")Use cases
What people build with it in Sanskrit.
Classes and lectures
Transcripts of Sanskrit teaching, live in the classroom or from the recording.
Audio and video to text ›Talks and videos
Captions for talks and recorded videos, and SRT files from their word timings.
Subtitles from video ›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.
Does Chariot have Sanskrit speech to text?
Yes. Set language=sa on a streaming session or a file, or leave the language out and the model detects it from the audio. It is one of the 12 languages in Chariot-1-ASR Realtime.
Can I try it on my own recordings first?
Yes. Every account starts with 10,000 free credits, almost two hours of audio, with no card. Upload a recording in the web app, or stream one over the API, and read the transcript before you decide anything.
Can I transcribe recorded lectures?
Yes. Upload a recording in Speech to Text in the web app, or send it to the file API. The async API takes recordings of up to three hours each, up to 100 in one batch.
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.
How is it billed?
1.5 credits per second of audio you send, in either mode, from the same credits as text to speech. Time with the connection open but no audio sent is not billed. Every session is saved to your workspace with its transcript and cost.
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.
Which languages does it support?
Hindi, Marathi, Bengali, Tamil, Telugu, Kannada, Malayalam, Gujarati, Punjabi, Odia, Sanskrit and English. Set the language for a session, or leave it out and the model detects it from the audio.
