Indic-Transcribe-flex

Multilingual speech recognition for 27 Indian languages — with native-script, mixed-script, and romanized output.

Quick links — Blog · Demo · Try it out · GitHub

Indic-Transcribe-pro is a multilingual Automatic Speech Recognition (ASR) model built for 27 Indian languages. It is trained to be robust and general purpose: it handles the full diversity of Indian accents and holds up in noisy real-world conditions, from crowded markets to call-center floors, with strong coverage in the domains where Indian voice products are actually built — education, agriculture, and healthcare.

Unlike traditional ASR systems that return only a native-script transcript, Indic-Transcribe offers three transcription modes — native script, romanized text, or true code-mixed output — so you can match the output to what your product expects.

This model is ready for commercial use.

---

Why Choose Indic-Transcribe?

• 🔀 Code-mixing, natively. Indians rarely speak one language at a time. Indic-Transcribe transcribes Hinglish and other mixed speech as it is actually spoken, instead of forcing it into a single language.
• 📝 Three transcription modes. Native script for fully native output; mixed script for native words in native script with English and numerals in Latin; romanized for everything in Latin script.
• 🏥 Domain coverage where it matters. Deep vocabulary in education, agriculture, and healthcare.
• 🌐 Language identification built in. Use the model directly as a language-ID system, or let it auto-detect the language and then transcribe.
• ⚡ Streaming for live use. Real-time transcription for conversational use cases, available in Indic-Transcribe-lite.

---

Supported Languages

The model covers 27 languages across four groups:

Group	Languages
Indian-accented English	English — benchmarked across speakers from 19 states
22 constitutionally recognised languages	Assamese, Bengali, Bodo, Dogri, Gujarati, Hindi, Kannada, Kashmiri, Konkani, Maithili, Malayalam, Manipuri, Marathi, Nepali, Odia, Punjabi, Sanskrit, Santali, Sindhi, Tamil, Telugu, Urdu
Hindi dialects	Bhojpuri, Chhattisgarhi, Haryanvi
Extremely low-resource	Bhili

See Supported Language Codes for the code to pass at inference time.

The three transcription modes

The same utterance, three renderings:

Mode	Output
Native script	मैंने कल पांच बजे तीन फाइलें अपलोड कीं
Mixed script	मैंने कल 5 बजे 3 files upload कीं
Romanized	maine kal 5 baje 3 files upload kin

---

Model Architecture

Architecture Type: NVIDIA Canary — FastConformer encoder with a Transformer decoder.

Indic-Transcribe-pro is built on the nvidia/canary-1b-v2 architecture. The FastConformer encoder produces acoustic representations that the Transformer decoder converts into text, with task tokens selecting the transcription mode and target language.

Model name	Indic-Transcribe-pro
Task	Speech-to-Text (Automatic Speech Recognition)
Base model	nvidia/canary-1b-v2
Total parameters	1.2B
Encoder	FastConformer — 32 layers, 811M params, 1024 hidden dim, 8 attention heads, conv kernel 9
Decoder	Transformer — 24 layers, 419M params, 1024 hidden size, 8 attention heads
Vocabulary	7,152 tokens (1,152 special / task + 6,000 multilingual)
Sub-word algorithm	BPE (byte fallback disabled)
Precision	fp16
Checkpoint size	4.6 GB

Feature comparison

Feature	Indic-Transcribe-flex	Indic-Transcribe-realtime
Languages	27	27
Code-mixed audio	✅	✅
Inverse Text Normalization	✅	❌
Transcription Modes	Native Script, Mixed Script, Romanized	Native Script
Automatic language ID	✅	❌
Streaming	❌	✅
Latency (1×H100)	150 ms	35 ms

---

Results at a Glance

Results below are from the Voice of India benchmark. ASR performance is measured using Word Error Rate (WER); lower is better. Best score per column is in bold.

Systems covering all benchmark languages

Model	Average	Assamese	Bhojpuri	Bengali	Gujarati	Hindi	Chhattisgarhi	Kannada	Maithili	Malayalam	Marathi	Odia	Punjabi	Tamil	Telugu	Urdu
Saaras V3	10.7	9.1	17.9	5.2	9.7	3.8	14.0	8.8	14.2	12.2	6.5	11.1	8.6	9.1	13.5	7.5
Indic-Transcribe-flex	11.3	9.6	18.5	5.0	10.8	4.1	13.6	9.7	15.1	13.7	6.6	10.2	9.5	10.8	13.1	5.6
Indic Conformer	17.8	13.1	30.3	9.5	16.3	6.5	24.5	16.3	16.3	28.2	11.6	13.1	19.1	16.2	20.0	8.0
Gemini 3 Pro	21.1	23.7	24.1	10.3	18.1	9.3	19.6	20.1	27.2	21.0	14.0	25.7	19.3	15.5	24.6	10.6
Gemini 3 Flash	23.1	25.0	20.2	11.1	20.3	7.1	21.7	18.5	28.0	30.1	14.2	23.3	23.2	17.3	25.5	11.0
Gemma E4B	36.1	45.0	27.0	19.6	27.3	9.1	24.0	31.0	36.7	44.4	24.4	44.4	23.4	37.9	41.6	14.2
OmniASR LLM 7B	44.5	23.9	26.3	20.9	32.0	9.6	20.7	35.0	44.6	48.8	24.5	72.3	31.7	40.6	48.7	14.8
OmniASR CTC 7B	62.5	33.3	39.2	43.5	62.5	20.7	33.4	50.0	51.6	61.3	33.6	90.2	80.7	58.3	62.1	89.3

---

How to Use this Model

Installation

pip install torch torchaudio transformers sentencepiece soundfile

The model code ships inside this repository, so there is nothing else to install — no NeMo, no other toolkit.

Conda environment (recommended for reproducibility)

conda create -n indic-transcribe python=3.10 -y
conda activate indic-transcribe
pip install torch torchaudio transformers sentencepiece soundfile

Input audio requirements

Sample rate	16 kHz (resampled automatically if it differs)
Channels	Mono
Formats	.wav, .flac, .mp3
Speakers	Single speaker — see Limitations

# Convert anything to the expected format
ffmpeg -i input.mp3 -ac 1 -ar 16000 -c:a pcm_s16le audio.wav

Basic inference

from huggingface_hub import snapshot_download
from indic_transcribe import IndicTranscribe

asr = IndicTranscribe.from_pretrained(snapshot_download("bodhan-ai/indic-transcribe-pro"))
print(asr("audio.wav", lang="hi"))

The three transcription modes

The same audio, three outputs. Pick the mode that matches what your downstream system expects.

Native script — everything in the language's own script. Recommended for production.

print(asr("audio.wav", lang="hi", mode="native"))
# मैंने कल पांच बजे तीन फाइलें अपलोड कीं

Mixed script (ITN) — native words in native script; English words and numerals in Latin. Inverse text normalization turns spoken numbers into digits. Recommended if you want formatted and normalized output.

print(asr("audio.wav", lang="hi", mode="mixed"))
# मैंने कल 5 बजे 3 files upload कीं

Romanized — everything transliterated into Latin script. Useful for search indexing, keyword spotting, and Latin-only UIs.

print(asr("audio.wav", lang="hi", mode="romanized"))
# maine kal 5 baje 3 files upload kin

Automatic language ID + transcription

When you don't know the language ahead of time, omit lang. The model identifies it first, then transcribes — at the cost of one decoder step, not a second encoder pass.

text, lid = asr.transcribe("unknown_language.wav", return_lid=True)
print(lid["lang"])   # e.g. "ta"
print(text)

A language you supply always wins; identification only fills a gap. return_lid=True also works when you did supply one, so a disagreement between your metadata and the model stays visible instead of silent:

text, lid = asr.transcribe("audio.wav", lang="hi", return_lid=True)
# lid == {"lang": "hi", "source": "explicit", "topk": [("hi", 0.9999), ("ur", 0.0001), ...]}

Language identification only

To use the model purely as a language-ID system, read the predicted language and discard the transcript.

for path in ["a.wav", "b.wav", "c.wav"]:
    print(path, asr.identify(path))
    # [('ta', 0.9812), ('ml', 0.0104), ('kn', 0.0031), ...]

identify returns the ranked distribution rather than a single string, because for the confusable pairs the top-1 alone hides how close the decision was. Accuracy is uneven: ml/ta 0.98 and kn/bn 0.96, against bho 0.05, hi 0.26, mai 0.36 and ur 0.49, each absorbed by a close neighbour. If you have a language label, pass it.

Streaming

Streaming is available in Indic-Transcribe-lite, not in pro. Use lite for live and conversational use cases where 30 ms latency matters more than peak accuracy.

---

Input(s)

Field	Details
Input Type(s)	Audio, Language ID
Input Format(s)	.wav, .flac, .mp3; string language code
Input Parameters	One-dimensional (1D) audio; one-dimensional (1D) language ID
Other Properties	16 kHz mono; audio is resampled automatically if it differs. Single speaker.

---

Output

Field	Details
Output Type(s)	Text string in the input language
Output Format(s)	String
Output Parameters	One-dimensional (1D)
Other Properties	Selectable native-script, mixed-script, or romanized rendering; optional detected-language tag.

---

Software Integration

Runtime Engine: NVIDIA NeMo (ASR collection)

Supported Hardware Microarchitecture Compatibility:

• NVIDIA Ampere
• NVIDIA Hopper
• NVIDIA Lovelace

Supported Operating System(s): Linux

---

Supported Language Codes

Pass these to source_lang. Use "auto" for automatic language identification.

Example	Value
Hindi	source_lang="hi"
Tamil	source_lang="ta"
Bengali	source_lang="bn"
Indian English	source_lang="en"
Auto-detect	source_lang="auto"

All 27 languages use standard ISO 639-1 / 639-3 codes (also listed in the language: field at the top of this card).

Full list of language codes (27)

Language	Code	Script	Group
English (Indian)	en	Latin	Indian-accented English
Assamese	as	Bengali–Assamese	Scheduled
Bengali	bn	Bengali	Scheduled
Bodo	brx	Devanagari	Scheduled
Dogri	doi	Devanagari	Scheduled
Gujarati	gu	Gujarati	Scheduled
Hindi	hi	Devanagari	Scheduled
Kannada	kn	Kannada	Scheduled
Kashmiri	ks	Perso-Arabic / Devanagari	Scheduled
Konkani	kok	Devanagari	Scheduled
Maithili	mai	Devanagari	Scheduled
Malayalam	ml	Malayalam	Scheduled
Manipuri	mni	Bengali / Meetei Mayek	Scheduled
Marathi	mr	Devanagari	Scheduled
Nepali	ne	Devanagari	Scheduled
Odia	or	Odia	Scheduled
Punjabi	pa	Gurmukhi	Scheduled
Sanskrit	sa	Devanagari	Scheduled
Santali	sat	Ol Chiki	Scheduled
Sindhi	sd	Perso-Arabic / Devanagari	Scheduled
Tamil	ta	Tamil	Scheduled
Telugu	te	Telugu	Scheduled
Urdu	ur	Perso-Arabic	Scheduled
Bhojpuri	bho	Devanagari	Hindi dialect
Chhattisgarhi	hne	Devanagari	Hindi dialect
Haryanvi	bgc	Devanagari	Hindi dialect
Bhili	bhb	Devanagari	Low resource

---

Limitations

• Single-speaker audio. The model is trained for single-speaker recordings. For multi-speaker scenarios, pair it with a diarization module and transcribe each speaker turn separately.
• Mixed-script (ITN) mode can be wrong. Inverse text normalization is genuinely ambiguous in places, so mixed-script mode will occasionally get it wrong. For production, native-script mode is recommended, applying your own ITN downstream if you need it.
• Romanization is not standardized. There is no single canonical romanization for most Indian languages. Romanized output follows the model's learned convention, which may not match your existing transliteration scheme.
• Streaming is not available in pro. Use Indic-Transcribe-lite for real-time use.
• Low-resource languages are weaker. Bhili in particular has very little training data. Check the per-language numbers in Results at a Glance before committing to a language.
• Noise robustness has limits. Heavily overlapped speech, far-field audio, and very low-SNR recordings will still degrade quality.

---

License / Terms of Use

Released under BODHAN AI OPEN MODEL LICENSE 1.0.

The base model, nvidia/canary-1b-v2, carries its own license terms — ensure your use complies with both.

---

Use Case

Transcription of multilingual and code-mixed Indian-language audio, including native-script, mixed-script, and romanized output, plus language identification.

---

Deployment Geography

Global

---

Citation

@misc{indictranscribe2026,
  title  = {Indic-Transcribe: Multilingual Speech Recognition for 27 Indian Languages},
  author = {TODO},
  year   = {2026},
  url    = {https://huggingface.co/bodhan/Indic-Transcribe-pro}
}

---

Ethical Considerations

The integration of foundation and fine-tuned models into AI systems requires additional testing using use-case-specific data to ensure safe and effective deployment. Developers should work with their team to ensure this model meets requirements for the relevant industry and use case, and addresses unforeseen product misuse.