Limit 1
The Tamil pack is conversational and may not match every regional register or a business that requires formal literary Tamil.
தமிழ் · measured India voice evidence
Tamil AI voice: runtime path, real pack lines and measured latency
Tamil resolves to Bharat Essential, Google Cloud Text-to-Speech voice ta-IN-Wavenet-A, the ta-IN dedicated pack and the cascade engine. Three fresh production-adapter runs measured a 857 ms median to the first non-empty 8 kHz PCM audio chunk on 15 Aug 2026.
Measured 15 Aug 2026 · three runs · 8 kHz PCM16 · no provider fallback
Which pack, engine and vendor voice serve Tamil?
The runtime trace starts with BaseBusiness.get_voice_pack, continues through voice_config.resolve_pack and ends at voice_config.resolve_engine. The generic integration returned none from the generic business integration, resolve_pack selected the ta-IN dedicated pack, and resolve_engine selected cascade. The TTS factory then selected Google Cloud Text-to-Speech voice ta-IN-Wavenet-A, displayed as Meena, while speech recognition remained Sarvam saaras:v3.
| Runtime check | Resolved value | Meaning |
|---|---|---|
| get_voice_pack | None from the generic business integration | A client-specific pack would override the generic registry result. |
| resolve_pack | ta-IN dedicated pack | Fixed lines and language rules come from a locale-specific pack. |
| voice_config | Bharat Essential · Google Cloud Text-to-Speech · ta-IN-Wavenet-A | The grade write-through pins the production TTS provider and voice identifier. |
| resolve_engine | cascade | The cascade engine uses separate STT, reasoning and TTS stages. |
| speech recognition | Sarvam saaras:v3 | The primary recognition locale is ta-IN. |
The runtime table names the actual Tamil call path rather than a language-support logo: generic integration result, resolved pack, sold grade, provider voice, engine and speech-recognition provider. A client integration can replace the generic pack, but the benchmark fixture used no client override.
What first-audio latency did Tamil measure?
Tamil measured 857, 863, 786 ms across three isolated production-adapter requests. The median was 857 ms. Each run synthesized “வணக்கம், நான் உங்களுக்கு எப்படி உதவலாம்?” at 8 kHz PCM16, started the timer immediately before speak(text), stopped at the first non-empty audio chunk, verified the configured adapter had not failed over, and cancelled the iterator after that first chunk.
| Run | First audio | Audio format | Adapter result |
|---|---|---|---|
| Run 1 | 857 ms | 8 kHz PCM16 | Google Cloud Text-to-Speech; no fallback |
| Run 2 | 863 ms | 8 kHz PCM16 | Google Cloud Text-to-Speech; no fallback |
| Run 3 | 786 ms | 8 kHz PCM16 | Google Cloud Text-to-Speech; no fallback |
The latency table reports a median of 857 ms from three measurements on 15 Aug 2026. The number is a reproducible TTS-adapter read, not end-to-end call latency, a percentile, or an uptime promise. Caller speech, transcription, model generation, carrier delivery and handset playback remain outside the timer.
What register does the resolved Tamil pack use?
The dedicated Tamil pack uses colloquial service Tamil. Endings such as -ங்க and forms such as உங்க keep the fixed lines spoken and respectful rather than formal written Tamil.
| Pack event | Exact runtime string | Register read |
|---|---|---|
| Filler while a tool runs | ஒரு நிமிஷம்-ங்க | Colloquial polite hold line with -ங்க. |
| Retry after unclear speech | மன்னிச்சுங்க, இன்னொரு தடவை சொல்லுங்க? | Spoken Tamil with polite plural endings. |
| Retract an unsupported action | மன்னிச்சுங்க, அத நான் இங்கிருந்து அனுப்ப முடியாது. உங்க நம்பர் குடுங்க, நம்ம டீம் உங்களுக்கு அனுப்பிடுவாங்க. | Conversational Tamil with a team handoff. |
The sample table copies three fixed strings from the ta-IN dedicated pack: the tool filler, unclear-speech retry and unsupported-action retraction. Tamil therefore has explicit runtime register evidence.
What are the honest Tamil limits?
Limit 2
The WaveNet adapter is non-streaming: Google returns the full clip before the first 200 ms PCM frame is yielded.
Limit 3
The three-run 857 ms median is a dated adapter benchmark rather than a carrier-to-handset SLA.
Limit 4
The first-audio benchmark measures TTS adapter latency only; it excludes caller speech recognition, model reasoning, carrier transit and handset playback.
Which evidence supports the Tamil hub?
The fresh page-level pull did not show a separate Tamil row in the selected 200-page output, but the existing Tamil post and the requested nine-language consolidation justify one non-combination hub. The existing language article remains the workflow and cost depth page; the multilingual category index provides cross-language buying context; and the product page describes the broader scheduled-language surface.
Other measured Indian-language hubs
The related hubs keep one route per language and never multiply languages by verticals. Every page publishes its own runtime trace, three dated latency reads and fixed-pack evidence.
Tamil voice questions
The answers preserve the runtime resolution, measured median, exact run values and pack limitation as standalone passages that do not depend on the tables.
Tamil resolves to Bharat Essential, the Google Cloud Text-to-Speech ta-IN-Wavenet-A voice, Sarvam saaras:v3 recognition and the cascade engine. The generic integration returned no client-specific pack, so resolve_pack selected the ta-IN dedicated pack on 15 Aug 2026.
Tamil measured 857 ms median first-audio latency across three fresh 8 kHz PCM16 adapter runs: 857, 863, 786 ms. The timer started immediately before the production adapter's speak call and stopped at the first non-empty audio chunk on 15 Aug 2026.
Tamil uses the ta-IN dedicated pack. The fixed pack strings shown on this page are copied from the runtime object rather than written as marketing examples.