Skip to content
Model infoModelsSnapshot: 15 Oct 2026

Addis AI / Addis speech synthesis

Addis Voices 2

Hosted speech synthesis with named voices, durable audio clips and HTTP/WebSocket streaming.

Speech & voice
availableClosed-source managed componentText-to-speech
Official documentation (opens in a new tab)Reviewed 16 Oct 2026
11 of 24 specification fields collected
Output formatsmp3_44100, wav_44100, pcm_16000 (WAV container)See specification notes below
Languages3 listedFull language list in specifications
Input billingSee ratesAddis AI
Output billingSee ratesPublished units and conditions below

Specifications

Provider-documented facts. “Not collected” means we have not verified that field yet, not that the feature is unsupported.

Context / input limit
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Maximum input tokens
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Maximum output
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Parameters
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Release date
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Knowledge cutoff
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Tool calling
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Structured output
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Streaming
SupportedSource for Streaming (opens in a new tab)

Framed HTTP audio or persistent WebSocket; these are TTS transports, not the conversation relay.

Batch processing
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Fine-tuning
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Reasoning
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Context caching
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Base model
Not collected

No model-specific disclosure reviewed; unknown is not unsupported.

Input modalities
TextSource for Input modalities (opens in a new tab)
Output modalities
AudioSource for Output modalities (opens in a new tab)
Languages
Amharic (am), Afaan Oromo (om), Tigrinya (ti)Source for Languages (opens in a new tab)

Main preview catalog has 28 am/om entries; newer streaming source has 35 including 7 ti voices.

Output formats
mp3_44100, wav_44100, pcm_16000 (WAV container)Source for Output formats (opens in a new tab)

HTTP stream accepts mp3_44100. WebSocket audio_format is mp3 or wav_mp3; inspect actual event format.

Voice controls
speed is applied. stability, similarity and style are accepted but ignored by the current provider and reported in ignoredVoiceSettings.Source for Voice controls (opens in a new tab)
Session duration
WebSocket: 10 minutes; one-use ticket valid 60 seconds; idle receive timeout 60 seconds.Source for Session duration (opens in a new tab)
Input limits
WebSocket cumulative text budget 1–5,000 characters set at creation; 100 request IDs/session; one generation at a time per connection; 20 messages/second. These are session limits, not a universal single-generation token limit.Source for Input limits (opens in a new tab)
Output duration
WebSocket audio/turn defaults to 60 seconds; max_audio_seconds can be set from 1 to 600.Source for Output duration (opens in a new tab)
Request quotas
No fixed session/request cap per API key is documented for this streaming service; wallet credit and language capacity bound use. 429 LANGUAGE_CAPACITY_REACHED includes retry delay.Source for Request quotas (opens in a new tab)
Safe request retries
Same client_request_id and identical inputs replay a completed generation without another charge. Changed text/voice/language/format requires a new ID.Source for Safe request retries (opens in a new tab)

Access & pricing

Prices belong to an endpoint and billing tier. A provider listing does not guarantee access for every account or region.

Addis AI

Provider source (opens in a new tab)
API model ID
Not collected

Provider model labels, backend selections and voice IDs do not establish an immutable request-side checkpoint selector.

Access
POST https://api.addisassistant.com/api/v1/voice/generations; x-api-key. text, voice_id, language, output_format and optional client_request_id. GET /api/v1/voice/voices supplies account availability.
Regions
Policy allows Ethiopia, Switzerland, EEA and provider regions. Exact request routing and private residency require agreement.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 16 Oct 2026
CategoryUsageRateUnit
Standard list priceList price1ETB 5.00generated audio minute
Dated launch promotionPromo through 2026-11-12 EAT2ETB 2.50generated audio minute

Rate conditions

  1. Native ETB list rate. Launch promotion applies through November 12, 2026 EAT; list rate applies again November 13. Final returned usage and account pricing control.
  2. 50% off for every account through the end of November 12, 2026 East Africa Time. Not a perpetual rate; no extra currency conversion. Standard/Turbo Scribe share the rate; billing uses final output characters.
Pricing source (opens in a new tab)

Data & enterprise terms

Standard-service terms/privacy permit content review, model training, fine-tuning and improvement. No-training or restricted processing requires a controlling written enterprise agreement. Account content and clips may persist for feature use, safety/legal needs or agreed terms; transient realtime processing has separately retained operational logs. No universal numeric retention TTL. Processing may occur in Ethiopia, Switzerland, EEA and provider regions; private residency requires written agreement.

Read the service policy (opens in a new tab)

Addis AI

Provider source (opens in a new tab)
API model ID
Not collected

Provider model labels, backend selections and voice IDs do not establish an immutable request-side checkpoint selector.

Access
POST https://api.addisassistant.com/api/v1/voice/generations/stream; x-api-key. Four-byte big-endian audio frame lengths; zero frame precedes final JSON. Handle JSON errors/replays before parsing audio.
Regions
Policy allows Ethiopia, Switzerland, EEA and provider regions. Exact request routing and private residency require agreement.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 16 Oct 2026
CategoryUsageRateUnit
Standard list priceList price1ETB 5.00generated audio minute
Dated launch promotionPromo through 2026-11-12 EAT2ETB 2.50generated audio minute

Rate conditions

  1. Native ETB list rate. Launch promotion applies through November 12, 2026 EAT; list rate applies again November 13. Final returned usage and account pricing control.
  2. 50% off for every account through the end of November 12, 2026 East Africa Time. Not a perpetual rate; no extra currency conversion. Standard/Turbo Scribe share the rate; billing uses final output characters.
Pricing source (opens in a new tab)

Data & enterprise terms

Standard-service terms/privacy permit content review, model training, fine-tuning and improvement. No-training or restricted processing requires a controlling written enterprise agreement. Account content and clips may persist for feature use, safety/legal needs or agreed terms; transient realtime processing has separately retained operational logs. No universal numeric retention TTL. Processing may occur in Ethiopia, Switzerland, EEA and provider regions; private residency requires written agreement.

Read the service policy (opens in a new tab)

Addis AI

Provider source (opens in a new tab)
API model ID
Not collected

Provider model labels, backend selections and voice IDs do not establish an immutable request-side checkpoint selector.

Access
POST /api/v1/realtime/sessions obtains one-use ticket; connect wss://api.addisassistant.com/api/v1/realtime/voice. Ticket valid 60 seconds for one connection.
Regions
Policy allows Ethiopia, Switzerland, EEA and provider regions. Exact request routing and private residency require agreement.

Prices retain their published currencies and billing units below.

Published billing units · original currency · observed 16 Oct 2026
CategoryUsageRateUnit
Standard list priceList price1ETB 5.00generated audio minute
Dated launch promotionPromo through 2026-11-12 EAT2ETB 2.50generated audio minute

Rate conditions

  1. Native ETB list rate. Launch promotion applies through November 12, 2026 EAT; list rate applies again November 13. Final returned usage and account pricing control.
  2. 50% off for every account through the end of November 12, 2026 East Africa Time. Not a perpetual rate; no extra currency conversion. Standard/Turbo Scribe share the rate; billing uses final output characters.
Pricing source (opens in a new tab)

Data & enterprise terms

Standard-service terms/privacy permit content review, model training, fine-tuning and improvement. No-training or restricted processing requires a controlling written enterprise agreement. Account content and clips may persist for feature use, safety/legal needs or agreed terms; transient realtime processing has separately retained operational logs. No universal numeric retention TTL. Processing may occur in Ethiopia, Switzerland, EEA and provider regions; private residency requires written agreement.

Read the service policy (opens in a new tab)

Caching

Caching can reuse earlier input processing. Availability, lifetime and billing depend on the endpoint. A cache hit still generates a new response.

01 / INITIAL INPUTProcess & retain

The system processes eligible input and retains reusable state.

02 / LATER REQUESTCheck available state

Reuse depends on matching input, session state and the cache policy.

03 / CACHE HITReuse & generate

Reusing state can reduce repeated work. Any charges follow the billing rules below.

Addis AI

Not verified

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

How it works
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Minimum reusable input
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Lifetime & refresh
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

How to enable it
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Usage fields
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Billing rules
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Limits & data boundaries
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Caching reviewed 16 Oct 2026

Addis AI

Not verified

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

How it works
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Minimum reusable input
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Lifetime & refresh
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

How to enable it
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Usage fields
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Billing rules
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Limits & data boundaries
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Caching reviewed 16 Oct 2026

Addis AI

Not verified

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

How it works
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Minimum reusable input
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Lifetime & refresh
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

How to enable it
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Usage fields
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Billing rules
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

Limits & data boundaries
Not collected

No native inference-cache contract reviewed. Request replay, durable clips, fileUri reuse and recovery windows are distinct from prompt caching.

No verified cache rate for this endpoint. An empty rate is not a zero charge.

Caching reviewed 16 Oct 2026

Voice choices

35 voices

Choose a voice within Addis Voices 2. These are API voice identifiers, not additional model releases. Source links document each voice.

Published voice catalog
Voice / API IDLanguage & accentDelivery
Hamenam-hamenAmharic
Not specified in the reviewed voice catalog		Warm conversational delivery; female; Conversational. Public documentation snapshot, live availability not authenticated.Voice source (opens in a new tab)
Simonam-simonAmharic
Not specified in the reviewed voice catalog		Calm storytelling voice; male; Commercial. Public documentation snapshot, live availability not authenticated.Voice source (opens in a new tab)
Yohannesam-yohanes-calmAmharic
Not specified in the reviewed voice catalog		Low-key reflective tone; male; Narration. Public documentation snapshot, live availability not authenticated.Voice source (opens in a new tab)
Tesfaam-tesfaAmharic
Not specified in the reviewed voice catalog		Confident, forward-leaning read; male; Commercial. Public documentation snapshot, live availability not authenticated.Voice source (opens in a new tab)
Muazam-muazAmharic
Not specified in the reviewed voice catalog		Clear expressive delivery; male; Commercial. Public documentation snapshot, live availability not authenticated.Voice source (opens in a new tab)
Robaam-robaAmharic
Not specified in the reviewed voice catalog		Energetic bright tone; male; Commercial. Public documentation snapshot, live availability not authenticated.Voice source (opens in a new tab)
Nejatam-nejatAmharic
Not specified in the reviewed voice catalog		Smooth measured elegance; female; Conversational. Public documentation snapshot, live availability not authenticated.Voice source (opens in a new tab)
Lozaam-lozaAmharic
Not specified in the reviewed voice catalog		Smooth polished narration; female; Narration. Public documentation snapshot, live availability not authenticated.Voice source (opens in a new tab)

What the measurements say

No benchmark observations have been added for this model. We do not turn a vendor description into a performance score.

Latency and throughput depend on the host, request size, reasoning settings and load. There is no measured speed or global usage figure in this record.

The record behind the model

Why it is included

Commercial hosted API with restricted service access and reserved provider/licensor IP. Public research artifacts exist; this entry is not an assertion of an independently closed-weight checkpoint.

Classification evidence (opens in a new tab)

Sources reviewed