Shunya LabsShunya LabsPlayground

Horizon-Odia

Odia Speech to Text for Real-World Conversations

Turn spoken Odia into searchable text for district-level service requests and radio archives. Shunya Labs Horizon Indic is designed for speech in Indian languages and regional varieties, with a platform for both uploaded recordings and live audio streams.

Streaming and batch transcriptionSpeaker labels and timestampsEnglish terms inside Odia speechSelf-hosted deployment
Odia audio in
Odia audioHorizon IndicOdia transcript
Search, LLM, voice agent, CRM, analytics, or automation
Language - OdiaModel - Horizon IndicStreaming APIBatch API

Odia speech to text

Why Odia Needs a Language-AwareSpeech Recognition Workflow

Odia is used in Odisha. A single clean sample rarely captures the practical demands of a deployed speech system. Real-world audio includes different microphones, spontaneous phrasing, local place names, names of people, interruptions and background noise. These details matter when a transcript feeds a support ticket, a community archive or a voice assistant.

For Odia, pay particular attention to Odia sounds, names of districts, and phone-quality audio. Select a reference script before testing, and compare the model output against transcripts checked by speakers familiar with the intended audience. A language name in a catalogue is the starting point for evaluation, not a guarantee that every dialect, script or audio channel performs identically.

Real Odia conversations may include

The audio a production system actually hears

  • Regional pronunciation
  • English business terminology
  • Customer names
  • Product names
  • Addresses
  • Numbers
  • Phone-quality audio
  • Background noise
  • Multiple speakers
  • Interruptions
  • Mixed-language vocabulary

Basic flow

Odia audioHorizon IndicOdia transcriptSearch / LLM / Voice agent / CRM / Analytics / Automation
CapabilityHorizon Indic
Odia speech recognition
Horizon Indic
Batch transcription
Real-time streaming
Automatic language detection
Speaker diarization
Speaker identification
Word-level timestamps
Confidence scores
Multilingual recognition
Code-Switch platform
Intent detection
Sentiment analysis
Emotion detection
Summarisation
Redaction
Streaming API
Batch API
Self-hosted deployment

Try it yourself

Test Real Odia Audio, Not Onlya Prepared Demo

Working with Odia audio requires attention to Odia sounds, names of districts, and phone-quality audio.Test your recordings, review the transcript in the writing system your audience expects,and connect the result to the workflows that matter to your team.

Test with

Audio from the workflow you are building

  • Customer call
  • Meeting
  • Interview
  • Voice note
  • Podcast
  • Sales call
  • Support conversation
  • Phone recording

Production testing should include

The words the business actually uses

  • Real names
  • Real addresses
  • Real product terms
  • Real business vocabulary
  • Real recording conditions

Horizon Indic for Odia

How Odia Speech-to-Text Works with Horizon Indic

Supply Odia audio through a supported file upload for batch transcription, or use a streaming connection for liveexperiences. Choose the deployed Indic model and, when supported, lock the correct language route; the publiclyreferenced code is “or” (confirm the code with the production language endpoint). Receive text to review,search, display in an application, or pass to an authorised downstream workflow. For supported batch configurations,consider diarization, timestamps and optional conversation intelligence. Feature availability can depend on configuration.

Same wider network

Odia sits with other supported Indian languages

  • Hindi
  • Bengali
  • Tamil
  • Telugu
  • Marathi
  • Kannada
  • Malayalam
  • Gujarati
  • Punjabi
  • Urdu
  • Assamese
  • Other supported Indian languages

Odia speech

Odia speechHorizon IndicOdia transcriptApplication

Odia-first transcription

Writing System and OdiaTranscript Quality

Odia is commonly associated with Odia script. Writing-system choices can still vary by community, context ordownstream application. Transcription captures the spoken language as text; transliteration changes the script;translation expresses the meaning in another language. These are distinct operations and should be testedseparately. Before deployment, confirm how the model represents Odia names, numbers, locally used terms andrelevant scripts. Keep a native-speaker-reviewed set of short recordings as a repeatable acceptance test.

Odia speech

Odia speechHorizon IndicOdia text

The transcript can then flow into

The systems that use the text

  • CRM
  • Search
  • Notes
  • Analytics
  • AI applications
  • Meeting records
  • Customer-support systems

Real-time Odia transcription

Convert Speech into Text whilethe User Is Still Talking

Horizon supports real-time speech recognition through WebSocket streaming. A live application does not need towait for the complete conversation. Partial text arrives while speech continues, then the segment is finalised.

Live audio

Microphone or application stream.

Live audioHorizonPartial transcriptFinal transcriptApplication

Partial and final transcripts

Audio startsPartial transcriptMore speech arrivesTranscript updatesFinal transcript segment

Best for

Interactive Odia applications

  • Odia voice agents
  • Live customer support
  • Live captions
  • Agent assist
  • Interactive AI
  • Dictation
  • Voice search

Streaming or batch

Batch vs Real-Time OdiaTranscription

Use streaming when text is needed now. Use batch when the full recording already exists.The speech-recognition goal stays the same. Only the processing mode changes.

Batch

Recorded Odia audioHorizonComplete transcriptOptional intelligence

Streaming

Odia happening now

Use real-time recognition when an application needs the user's Odia words during a conversation - for example a live caption screen, a voice-driven form or agent assistance. Shunya documents WebSocket streaming in its Horizon platform. Validate latency, audio quality and per-language recognition using the deployed service rather than assuming an aggregate benchmark applies to Odia.

  • Odia voice agents
  • Live customer support
  • Live captions
  • Agent assist
  • Interactive AI
  • Dictation
  • Voice search

Batch

Odia already recorded

Use batch transcription when you already have an Odia interview, phone recording, podcast, lecture or field recording. The system processes the completed audio and returns text that reviewers can correct and tag. This mode is a useful starting point for archives and back-office processing because reviewers can inspect the entire recording.

  • Recorded customer calls
  • Meetings
  • Interviews
  • Podcasts
  • Voice notes
  • Research
  • Media
  • Post-call analysis
RequirementStreamingBatch
Live Odia speech
Recorded Odia audio
Partial transcripts
Complete transcript
Voice agents
Post-call intelligence
WebSocket
HTTP API

Language and model

Detect the Language, Then Choosethe Path

Horizon supports automatic language detection across its multilingual speech platform. When an application alreadyknows the user is speaking Odia, set the deployed configuration accordingly. Horizon Indic lists Odia as languagecode or. Use the code the current language listing exposes for the deployment you are calling.

Automatic language detection

Speech arrivesHorizonLanguage detectedTranscript

01

Odia-first speech

Use Horizon Indic.

02

Mixed-language speech

Use the appropriate Code-Switch model where the required pair is supported.

03

Clinical speech

Use the appropriate domain-specific speech model where it applies.

The conversation determines the recognition path

Indian-language speech → IndicMixed-language speech → Code-SwitchClinical speech → domain model

Shunya supports different Horizon model families for different speech environments.

Odia and English vocabulary

Real Business Conversations CanContain Mixed Terminology

Odia speakers may use English terms inside an otherwise Odia conversation. Horizon includes Code-Switchmodels for speech that moves between languages in the same interaction. A dedicated Odia-English pairVerify Odia-English Code-Switch routing in your active deployment and test with your own mixed audio.

English terms that show up in Odia speech

Business vocabulary

  • Order
  • Payment
  • Account
  • Support
  • Delivery
  • Appointment
  • Login
  • Status
  • Meeting
  • Report

Safe implementation

Real Odia-English audioCheck active deployment supportChoose the supported recognition routeEvaluate the actual transcript

Check the supported deployment pair, use that recognition route, and evaluate the transcript on real mixed audio before you rely on it.

Names, numbers, and references

Example: From Odia Audioto Actionable Text

A local organisation records an Odia service enquiry and needs a readable transcript before responding.Input: a consented Odia recording containing a real enquiry, interview or instruction.ASR: select Horizon Indic and review the transcript with an Odia speaker, checking names, dates and the expected script.Optional translation: send the reviewed transcript to Parallax only if the exact language pair has been validated.Output: a searchable record for a CRM, media archive, support dashboard, content workflow or educational resource.

Proper nouns

Names and places in Odia recordings

  • Customer names
  • Company names
  • Product names
  • Villages
  • Cities
  • Hospitals
  • Addresses
  • Brands

Operational details

Numbers, dates, and references

  • Order numbers
  • Transaction IDs
  • Phone numbers
  • Dates
  • Appointment times
  • Amounts
  • Quantities

Production test

Include the words customers actually say

  • Real customer names
  • Actual business terms
  • Real places
  • Product terminology
Consented Odia recordingHorizon Indic + Odia speaker reviewOptional Parallax translationSearchable record

Illustrative workflow only: no generated native-language sentence or translation here is presented as a measured product result.

Who said what

Separate People, Then ResolveKnown Participants

Speaker diarization assigns transcript segments to different speakers. Speaker identificationcan resolve registered voice profiles where it is configured. Unknown speakers stayon anonymous labels. Identification is not legal identity verification.

Diarization

Speaker-labelled segments

SPEAKER_00 Customer request
SPEAKER_01 Agent response
SPEAKER_00 Customer follow-up

Before

Anonymous label

SPEAKER_01 Agent response

After

Registered profile

Agent Priya Agent response

Use diarization for

Conversations with more than one voice

  • Customer calls
  • Meetings
  • Interviews
  • Podcasts
  • Sales conversations
  • Research

Timing and confidence

Connect Each Word to the Audio, andFlag What Needs Review

Verbose Horizon output can include a word start time, a word end time, and a confidence score.Confidence is an operational signal. It is not proof that a transcript word is correct.

Word-level timestamps

Jump from a word back to the recording

  • Search
  • Meeting playback
  • Call review
  • Captions
  • Subtitles
  • Media editing
  • Transcript highlighting

Confidence scores

Where a reviewer should look first

  • QA
  • Name checking
  • Number verification
  • Search ranking
  • Manual review

Search

KeywordFind transcript matchUse timestampJump to the corresponding audio

Where Odia speech becomes text

Practical Odia Use Cases

Voice-Product Developers

Enable an in-app listening layer so people can interact with services without typing lengthy messages. The immediate opportunity for Odia is district-level service requests and radio archives.

Operations & Contact Centres

Connect transcripts to case management, QA review, agent assistance and downstream analytics. This is especially useful when the source material contains Odia sounds, names of districts, and phone-quality audio.

Public Access & Inclusion

Turn spoken questions into reviewable text for community information desks and public-facing applications. Keep human review in the workflow until your acceptance thresholds have been met.

Customer support

Customer callHorizonOdia transcriptSupport applicationSearch / CRM / QA / Analytics

Meetings

MeetingHorizonTranscriptSpeaker labelsWord timingSearch / Notes / AI / Knowledge base

Odia voice agents

User speaks an Odia requestHorizon returns the Odia transcriptLLM or agent understands the requestRadiance generates the spoken replyUser hears the response

Phone and file audio

Use the Recordings your WorkflowAlready Creates

Odia customer conversations may arrive over ordinary phone systems. For video, the recognitionmodel follows the spoken audio rather than the video container.

Live stream

Documented streaming formats

  • int16 PCM
  • float32
  • μ-law
  • A-law

Batch

Documented file formats

  • WAV
  • MP3
  • M4A
  • OGG
  • FLAC
  • WebM

Phone and telephony

Live calls

  • Contact centers
  • Customer support
  • Voice agents
  • Agent assist

Files

Where the recordings come from

  • Mobile recordings
  • Meetings
  • Podcasts
  • Voice notes
  • Customer calls
  • Browser workflows

Phone call

Phone callLive audio streamHorizonOdia transcript

Recorded file

Recorded Odia audioHorizonTranscript

Video

VideoAudio trackHorizonTranscriptCaptions / Search / Editing / Analytics

Post-call intelligence

Turn a Transcript into StructuredConversation Data

For supported batch workflows, Horizon can add speakers, timing, intent, sentiment, emotion,a summary, and redaction. These enrichments are positioned as batch and post-callcapabilities. Sentiment supplements what the speaker said. It does not replace it.

Diarization

Who spoke in each segment.

Speaker identification

Resolve registered voice profiles where configured.

Intent

Why the conversation happened, using the organisation's own taxonomy.

Sentiment

A contextual signal for QA, experience analysis, escalation review, coaching, and research.

Emotion

An inferred post-call signal, not a verified fact about the speaker.

Summary

Reason for contact, issue, information, action, escalation, and next step.

Word timing

Connect the transcript back to the recording.

Redaction

Mask configured sensitive terms. One layer of a wider data-governance strategy.

Intent

The organisation defines the taxonomy

  • Order status
  • Payment query
  • Appointment
  • Customer support
  • Product enquiry
  • Human-agent request

Batch today

Odia audioTranscriptSpeakers + timeOptional intelligenceCRM / Search / Analytics / LLM

Summarisation

RecordingOdia transcriptSummaryCRM / Support / Analytics

Redaction

Original transcriptConfigured redactionMasked output

Call archives, CRM, QA, search, and analytics.

Developers

Developer Integration Notes

Use the Horizon ASR API for speech recognition and Parallax for text translation. For Odia, the editorial language identifier is or; obtain the accepted code and model ID from the production API. Shunya's current main documentation calls the Indic model horizon-indic, while older documentation may refer to zero-indic.

odia-transcription.json

{
  "file": "odia-call.wav",
  "model": "horizon-indic",
  "language_code": "or"
}

Configure as needed

Recommended configuration checklist

  • Authentication
  • Supported audio format and sampling rate
  • Model ID
  • Explicit language selection versus auto-detect
  • Required script
  • Timestamps
  • Speaker labels
  • Error handling
  • Retention and access rules

Output can include

Structured transcript

  • Transcript text
  • Segments
  • Detected language
  • Audio duration
  • Timing
  • Speakers
  • Enabled intelligence fields

Basic flow

Odia audioHorizon IndicStructured transcriptYour application
Explore the Speech to Text API

Benchmarks

Test Odia accuracy on your ownaudio

Broader Horizon evaluations include Hindi and other platform results. They are not an Odia word error rate, an Odia accuracypercentage, or an Odia benchmark rank. A real production sample is more meaningful than an unsupported number.

Test it on your own Odia audio before production.

Production testing

Evaluate with the Odia audio you will actually process

  • Real Odia customer calls
  • Regional speech
  • Business vocabulary
  • Names
  • Numbers
  • Phone audio
  • Multiple speakers
Test Horizon on your Odia audio

Built for production

Beyond One Odia Transcription Sample

A production Odia workflow may need recognition, streaming, batch processing, speaker structure,intelligence, an API, and room to add other Indian languages. Self-hosted deployment is part of thebroader production feature set where an organisation needs more control over infrastructure.Follow current Shunya deployment documentation for the exact configuration.

Self-hosted deployment

Where infrastructure control matters

  • Data residency
  • Internal infrastructure
  • Controlled speech-processing environments
  • Deployment governance
RequirementHorizon
Odia speech to text
Horizon Indic
Batch transcription
Streaming transcription
Automatic language detection
Multilingual recognition
Code-Switch platform
Speaker diarization
Speaker identification
Word timestamps
Confidence scores
Telephony workflows
Intent detection
Sentiment analysis
Emotion detection
Summarisation
Redaction
API integration
Multilingual expansion
Self-hosted deployment
Odia speechHorizonStructured text

Structured text

Ready for the systems that use it

  • Search
  • CRM
  • Analytics
  • LLM
  • Voice agent
  • Knowledge base

Questions

Frequently Asked Questions

Can Shunya Labs transcribe Odia audio?

Odia appears in the published Horizon Indic supported-language list. Test the current API with your own speakers, scripts, accents and recording conditions before production use.

Can I transcribe Odia in real time?

Horizon offers both streaming and batch workflows. Verify the exact Odia configuration and performance in the target deployment.

What writing system does Odia speech-to-text use?

Odia is associated with Odia script. Confirm that the deployed output format matches your audience and do not assume every alternate script is automatically available.

Can Odia audio be translated into English?

The normal route is speech transcription followed by Parallax text translation. The published evaluation reports 39 BLEU, but real-world content and direction still require evaluation.

How should I evaluate Odia transcription accuracy?

Use consented recordings that represent your actual speakers and tasks. Review proper nouns, Odia sounds, names of districts, and phone-quality audio, and output-script consistency rather than relying only on a multilingual platform-wide benchmark.

Get started

Build Your Odia VoiceWorkflow

Try real Odia audio in the Shunya Labs playground, inspect the output, and speak with the team about streaming,APIs and translation support. Bring the scripts, speaker groups, audio samples and required downstreamlanguage that your application will use.