Sarvam AI
Sarvam Motif

Speech to Text, perfected for Bengali

Every accent. Conjunct consonants. Benglish switching. Saaras hears Bengali the way Bengalis actually speak it.

Speak Bengali to see live captions

Every Bengali word, captured accurately

Code-mixing. Numbers. Proper nouns. Abbreviations.
0:00
হ্যাঁআমিPROPER NOUNFlipkartথেকেNUMBER₹12,499এরএকটাENTITYSamsungGalaxyCODE SWITCHphoneorderকরেছিলাম।OrderIDহলোABBREVIATIONFLK-78234-XN।CODE SWITCHDeliveryaddressহলো42,PROPER NOUNKoramangala5thBlock,PROPER NOUNBangalore।আমারphonenumberহলোPHONE NUMBER9840950950।CODE SWITCHPleasedeliveryটাএকটুতাড়াতাড়িকরেদিন,আমারপুরোনোফোনটাএকদমকাজকরছেনা।

Flexible output. Same Bengali audio,
multiple formats

Transcribe

With formatting and number normalization.

Output
আমার ফোন নম্বর হলো 9840950950

Translate

From Indic languages to English.

Output
My phone number is 9840950950

Transliteration

Indian languages written in English letters.

Output
amar phone number holo 9840950950

Verbatim

Preserves fillers and spoken numbers.

Output
আমার ফোন নম্বর হলো নয় আট চার zero নয় পাঁচ zero নয় পাঁচ zero

Saaras hears বাংলা the way you speak it

Every feature shown with real Bengali audio. Play each clip. Read the transcript. See the difference.

Speaker Diarization

Know who said what

Automatically labels speakers in multi-party Bengali conversations. Colors match the transcript.

Click to play
0:04Speaker 1

নমস্কার, HDFC bank-এ call করার জন্য ধন্যবাদ। আমি প্রিয়া বলছি, আপনাকে কীভাবে help করতে পারি?

0:09Speaker 2

হ্যাঁ, আমি আমার savings account-এর balance check করতে চাই। Account number হলো 2641-0078-3395।

0:18Speaker 1

অবশ্যই, verification-এর জন্য আপনার registered mobile numberটা confirm করুন।

Code-Mixing

Seamless Benglish

Handles mid-sentence switching between Bengali and English. No drops, no accent shifts.

Click to play
আজকের Code-switchmeetingটা খুব productive ছিল। Lunch-এর পর Englishcoding session-টা একদম smooth গেছে। Team দারুণ কাজ করেছে।
Telephony-Optimized

Built for real calls

8kHz call center audio, compression, background noise — maintains accuracy for Bengali where others fail.

8kHzNoisy
Click to play

আচ্ছা, বর্তমান সরকার-এর কাজকর্ম আপনার কেমন লাগছে? দেখুন, রাস্তাঘাট তো হয়েছে, কিন্তু জিনিসপত্রের দাম অনেক বেড়ে গেছে। পেট্রোলের দাম ₹108 হয়ে গেছে। সাধারণ মানুষ একটু relief চায়। তবে metro-র কাজ ভালো চলছে, Nagpur-এ অনেক development হচ্ছে।

Developer-first platform

OpenAI-compatible APIs. Drop-in SDKs for Python and Node.js. Go from zero to first transcription in under 5 minutes.

REST & WebSocket APIs

Standard REST for batch, WebSocket for real-time streaming with sub-250ms first byte.

SDKs & libraries

Official Python and Node.js SDKs with TypeScript support. pip install sarvam-ai.

Complete documentation

Interactive API reference, code samples, and integration guides for every endpoint.

Free tier included

Start building immediately. No credit card, no sales call, no minimum commitment. 

from sarvamai import SarvamAI

client = SarvamAI(api_subscription_key="YOUR_SARVAM_API_KEY")

# Transcribe audio
response = client.speech_to_text.transcribe(
    file_path="call_recording.wav",
    language_code="hi-IN",
    model="saaras:v3",
    with_diarization=True
)

print(response.transcript)
for turn in response.turns:
    print(f"[{turn.speaker}] {turn.text}")

Enterprise-ready. Responsible AI.

Built with safety, compliance, and data sovereignty at the core.

SOC 2 Type II & ISO 27001

Enterprise-grade security certifications. Annual audits, documented controls, continuous monitoring.

No training on your data

Your API inputs are never used for model training. Zero data retention after processing unless explicitly requested.

Data sovereignty

All data processed and stored in India. No cross-border transfers. Full compliance with Indian data regulations.

Audio data privacy

Audio data is processed in real-time and never stored unless explicitly requested. Complete privacy by default.

Content safety filters

Automated detection and filtering of harmful, abusive, or misleading content in transcriptions.

Audit-ready logging

Comprehensive API usage logs, access controls, and RBAC for enterprise governance and compliance reporting.

Simple, transparent pricing

Start free. Scale as you grow. No hidden costs.

Base plan

₹1.25 per minute

Free trial included

No credit card required. Get API keys instantly.

Volume discounts available
Enterprise pricing available
Speaker diarization included
Usage analytics
All 22 languages
Best for startups

Your questions, answered
Saaras v3 achieves industry-leading accuracy for Bengali, correctly processing complex conjunct consonant forms across literary and colloquial speech.
Yes. Saaras v3 handles natural Bengali-English code-switching seamlessly without accent shifts or dropped words.
Yes. Our WebSocket streaming API delivers sub-250ms latency for Bengali, ideal for real-time voice agents and conversational AI.
Saaras v3 is trained on diverse Bengali speech data including Kolkata Bengali, Bangladeshi Bengali influences, and regional variations across West Bengal.
10+ formats including MP3, WAV, AAC, OGG, Opus, FLAC, M4A, AMR, WMA, and WebM. Streaming API supports WAV and raw PCM at 16kHz.

Transcribe in 22 Indian languages Powered by Saaras v3