falaai-api 1.21.51 → 1.21.54

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: d83d046b979cd6c1d09c21c3c76a82ea6b0d17131e1018b972a7fea0a3a50f3e
-  data.tar.gz: ba6256865b059b177181ce24f00cd857a2df99b0af69584fd8b4131b3a00d7df
+  metadata.gz: 2124fbb4e1ed6511a5749326aa954da95d70c298e244a5b80dc09c214b66f8d3
+  data.tar.gz: ea43b97e08f90260511970b2616066aef0433d0948a512a12e34c36bfa7ba0e2
 SHA512:
-  metadata.gz: 74c598532ababe5e246b16f3e260886235f8b5e7f25df353b3d7cd2dc27ee392c90dbd8b1d0fac9ece144e7f04efa1209368c102c847e5008e476eecf1a8c8be
-  data.tar.gz: d3590c7c1c2e1e20bb4157f73125daae0c297dff4d44b985f3fcb7df100f4233a3672e9e200e2f68592fa61b0cf804c2099bfeb2f7b3fc815221bf738aea537a
+  metadata.gz: 9c41a3551e9c471ff24cdfbd1849ff772af1ed55844834dfb22299d0a6f5f44e5bd7b16f2ac3595db8df0a02d7c62771c53fd0b02982351f883cec089dfe5808
+  data.tar.gz: c90657bb08a24f46ca3854bea50be4f6e1a05e60885c3afab871f51925fc238b2b98565456a43d11aa5e3d945081f217f704a9dc1d2696e8eed8b0494d303627

data/README.md

@@ -123,6 +123,7 @@ FalaAI API is an **AI conversation-intelligence API** for analyzing customer-ser
 - **Contact center / Quality** - audit 100% of conversations instead of a sample.
 - **Compliance / Legal** - auditable evidence for audits and disputes.
 - **CX / Operations** - risk score, sentiment and reason per conversation.
+- **Sales** - analyze sales calls and extract structured outcomes.
 - **BI / Data** - typed JSON ready for your database or analytics stack.
 
 ## Why FalaAI
@@ -131,7 +132,7 @@ FalaAI API is an **AI conversation-intelligence API** for analyzing customer-ser
 - **Auditable by design** - a compliance risk score with detections/violations and an HTML report you can hand to an auditor.
 - **Credits, not tokens** - you know exactly what each call costs (per second of audio, per character of text, per conversation). Monthly plans + non-expiring top-up packs.
 - **Zero-storage** - audio/text/results are processed and discarded; only usage/audit records remain. TLS in transit + at rest, per-account isolation (multitenancy), LGPD (you are the controller; Action Tec is the processor).
-- **3 native languages** (PT/EN/ES) · **start free** (no card).
+- **Multilingual** — transcription in **31 languages**, diagnosis in **21 locales**, risk audit and HTML reports in **3** (PT/EN/ES) · **start free** (no card).
 - **B2B ready** - one Enterprise plan serves N clients, no per-user fee (ideal for ISVs).
 
 ## Integrations
@@ -202,7 +203,7 @@ end
 | Requirement | Version |
 |---|---|
 | Ruby | 2.7+ |
-| API | v1.21.51 |
+| API | v1.21.54 |
 
 ## Documentation
 
@@ -215,7 +216,7 @@ end
 
 ## Versioning
 
-Semantic versioning; the SDK version tracks the API version (`1.21.51`). See [CHANGELOG.md](CHANGELOG.md) and [Releases](https://github.com/ActionTecBr/falaai-api-ruby/releases).
+Semantic versioning; the SDK version tracks the API version (`1.21.54`). See [CHANGELOG.md](CHANGELOG.md) and [Releases](https://github.com/ActionTecBr/falaai-api-ruby/releases).
 
 ## Security
 

data/docs/AnalysisApi.md

@@ -26,7 +26,7 @@ FalaAI.configure do |config|
 end
 
 api_instance = FalaAI::AnalysisApi.new
-diagnostic_request = FalaAI::DiagnosticRequest.new({language: 'language_example', duration_seconds: 3.56}) # DiagnosticRequest | 
+diagnostic_request = FalaAI::DiagnosticRequest.new({language: 'language_example'}) # DiagnosticRequest | 
 
 begin
   # Analyze a call transcript — 5 parallel analyses
@@ -93,7 +93,7 @@ FalaAI.configure do |config|
 end
 
 api_instance = FalaAI::AnalysisApi.new
-risk_audit_request = FalaAI::RiskAuditRequest.new({duration_seconds: 3.56, language: 'language_example', response_language: 'response_language_example'}) # RiskAuditRequest | 
+risk_audit_request = FalaAI::RiskAuditRequest.new({dialog: 'dialog_example', duration_seconds: 3.56, language: 'language_example', response_language: 'response_language_example'}) # RiskAuditRequest | 
 
 begin
   # Compliance Risk Audit — conversation compliance analysis

data/docs/DiagnosticParticipant.md

@@ -6,7 +6,7 @@
 | ---- | ---- | ----------- | ----- |
 | **interlocutor** | **String** | Exact speaker label from the dialog (e.g. 'Speaker 1') |  |
 | **name** | **String** | Participant name (optional) | [optional] |
-| **role** | **String** | agent | client | bot |  |
+| **role** | **String** | agent | client | bot (optional — omit/empty when unknown) | [optional] |
 
 ## Example
 

data/docs/DiagnosticRequest.md

@@ -10,8 +10,8 @@
 | **audio_events** | [**Array<DiagnosticAudioEvent>**](DiagnosticAudioEvent.md) | Detected audio events with timestamps (required when using dialog) | [optional] |
 | **language** | **String** | Transcript language. Required. Accepted: en-US, pt-BR, es-ES, es-MX, fr-FR, de-DE, it-IT, pt-PT, zh-CN, ja-JP, ko-KR, ar-SA, hi-IN, ru-RU, id-ID, tr-TR, nl-NL, pl-PL, vi-VN, th-TH, en-GB |  |
 | **response_language** | **String** | Language for the analysis results (labels, categories, levels). Can differ from 'language' (input). If omitted, uses 'language'. Accepted: en-US, es-ES, es-MX, pt-BR, fr-FR, de-DE, it-IT, pt-PT, zh-CN, ja-JP, ko-KR, ar-SA, hi-IN, ru-RU, id-ID, tr-TR, nl-NL, pl-PL, vi-VN, th-TH, en-GB. | [optional] |
-| **duration_seconds** | **Float** | Total audio duration in seconds. Required. Max 3h (10800s). |  |
-| **client_reference_id** | **String** | Optional client-supplied ID echoed verbatim in the response. Use to correlate/sync with your system. Accepted charset: [A-Za-z0-9._:-], max 128 chars. Not idempotency. | [optional] |
+| **duration_seconds** | **Float** | Total audio duration in seconds. Optional when using only 'text' (ignored — not read/used). Required when using 'dialog' (must be >= last turn end). Max 3h (10800s). | [optional] |
+| **client_reference_id** | **String** | Optional client-supplied ID echoed verbatim in the response. Use to correlate/sync with your system. Accepted charset: [A-Za-z0-9 _/:.-], max 60 chars. Not idempotency. | [optional] |
 | **call_direction** | **String** | Who originated the call. inbound=client called, outbound=company called. If omitted, the LLM infers from context. | [optional] |
 | **participants** | [**Array<DiagnosticParticipant>**](DiagnosticParticipant.md) | Explicit participant roles. If omitted, the LLM infers from the dialog. When provided, used as ground truth (no inference) and echoed in analysis.participants_identified. | [optional] |
 

data/docs/DiagnosticResponse.md

@@ -10,6 +10,7 @@
 | **analysis** | [**DiagnosticAnalysisMap**](DiagnosticAnalysisMap.md) | The 6 conversation analyses (5 + participants) |  |
 | **usage** | [**DiagnosticUsage**](DiagnosticUsage.md) | Usage and processing information |  |
 | **client_reference_id** | **String** | Client-supplied ID echoed verbatim (if provided in request) | [optional] |
+| **html_report** | **String** | Visual HTML report of the diagnosis (base64 gzip). Only present when 'dialog' is provided; null for text-only requests. | [optional] |
 
 ## Example
 
@@ -22,7 +23,8 @@ instance = FalaAI::DiagnosticResponse.new(
   object: null,
   analysis: null,
   usage: null,
-  client_reference_id: null
+  client_reference_id: null,
+  html_report: null
 )
 ```
 

data/docs/Participant.md

@@ -6,7 +6,7 @@
 | ---- | ---- | ----------- | ----- |
 | **interlocutor** | **String** | Exact identifier as used in dialog (e.g. 'Interlocutor 1', 'Antonio') |  |
 | **name** | **String** | Participant name (humanizes report, does not affect logic) | [optional] |
-| **role** | **String** | Role: agent (human operator), client (customer), bot (IVR/AI) |  |
+| **role** | **String** | Role: agent (human operator), client (customer), bot (IVR/AI). Required for risk audit. |  |
 
 ## Example
 

data/docs/ParticipantDiagnostic.md

@@ -5,7 +5,7 @@
 | Name | Type | Description | Notes |
 | ---- | ---- | ----------- | ----- |
 | **interlocutor** | **String** | Exact speaker label from the dialog (e.g. 'Speaker 1') |  |
-| **role** | **String** | Role: agent | client | bot | agent_requester | agent_custodian |  |
+| **role** | **String** | Role: agent | client | bot | agent_requester | agent_custodian (null when not informed/inferred) | [optional] |
 | **name** | **String** | Participant name if mentioned in the dialogue | [optional] |
 | **confidence** | **String** | high | medium | low | [optional] |
 | **evidence** | **String** | Exact verbatim quote supporting the role (no timestamps) | [optional] |

data/docs/RiskAuditRequest.md

@@ -5,16 +5,15 @@
 | Name | Type | Description | Notes |
 | ---- | ---- | ----------- | ----- |
 | **model** | **String** | Analysis model. Always 'falaai-risk-audit-1' | [optional][default to 'falaai-risk-audit-1'] |
-| **text** | **String** | Plain transcript (fallback if dialog is empty). At least one of 'dialog' or 'text' required. Max 300,000 characters | [optional][default to ''] |
-| **dialog** | **String** | Diarized transcript with speaker turns. PRIMARY source. Speaker labels accepted (any case): 'Speaker N', 'Interlocutor N', 'Hablante N', 'Locutor N', 'Orador N' (space or underscore). Normalized internally to 'Speaker N' in the response. Max 300,000 characters | [optional][default to ''] |
+| **dialog** | **String** | Diarized transcript with speaker turns. REQUIRED — risk audit only accepts diarized conversations; plain 'text' is not accepted. Speaker labels accepted (any case): 'Speaker N', 'Interlocutor N', 'Hablante N', 'Locutor N', 'Orador N' (space or underscore). Normalized internally to 'Speaker N' in the response. Max 300,000 characters |  |
 | **audio_events** | [**Array<DiagnosticAudioEvent>**](DiagnosticAudioEvent.md) | Audio events with timestamps (correlated with turns when diarization is present) | [optional] |
-| **duration_seconds** | **Float** | Total audio duration in seconds. Required. Max 3h (10800s). |  |
+| **duration_seconds** | **Float** | Total audio duration in seconds. Required. Must be >= last turn end. Max 3h (10800s). |  |
 | **language** | **String** | Language of the transcript being analyzed. Must match the dialog/text language. Accepted: pt-BR, en-US, es-ES. |  |
 | **response_language** | **String** | Language for analysis results (labels, categories, levels, actions, HTML report). Can differ from 'language'. Accepted: pt-BR, en-US, es-ES. |  |
 | **call_direction** | **String** | Who originated the call. inbound=client called, outbound=company called. If omitted, LLM infers from context. | [optional] |
 | **participants** | [**Array<Participant>**](Participant.md) | Explicit participant roles. If omitted, LLM infers from dialog (Lei 17). When provided, used as ground truth — no inference. | [optional] |
 | **response_format** | **String** | Response format version. Only 'v2' (structured EN-US blocks) is available today. | [optional][default to 'v2'] |
-| **client_reference_id** | **String** | Optional client-supplied ID echoed verbatim in the response. Use to correlate/sync with your system. Accepted charset: [A-Za-z0-9._:-], max 128 chars. Not idempotency. | [optional] |
+| **client_reference_id** | **String** | Optional client-supplied ID echoed verbatim in the response. Use to correlate/sync with your system. Accepted charset: [A-Za-z0-9 _/:.-], max 60 chars. Not idempotency. | [optional] |
 
 ## Example
 
@@ -23,7 +22,6 @@ require 'falaai-api'
 
 instance = FalaAI::RiskAuditRequest.new(
   model: null,
-  text: null,
   dialog: null,
   audio_events: null,
   duration_seconds: null,

data/docs/SpeechApi.md

@@ -29,7 +29,7 @@ file = File.new('/path/to/some/file') # File |
 opts = {
   model: 'model_example', # String | 
   language: 'language_example', # String | 
-  client_reference_id: 'client_reference_id_example' # String | Optional client-supplied ID echoed verbatim in the response. Use to correlate/sync with your system. Accepted charset: [A-Za-z0-9._:-]. Not idempotency.
+  client_reference_id: 'client_reference_id_example' # String | Optional client-supplied ID echoed verbatim in the response. Use to correlate/sync with your system. Accepted charset: [A-Za-z0-9 _/:.-], max 60 chars. Not idempotency.
 }
 
 begin
@@ -65,8 +65,8 @@ end
 | ---- | ---- | ----------- | ----- |
 | **file** | **File** |  |  |
 | **model** | **String** |  | [optional][default to 'falaai-transcribe-1'] |
-| **language** | **String** |  | [optional][default to 'pt'] |
-| **client_reference_id** | **String** | Optional client-supplied ID echoed verbatim in the response. Use to correlate/sync with your system. Accepted charset: [A-Za-z0-9._:-]. Not idempotency. | [optional] |
+| **language** | **String** |  | [optional] |
+| **client_reference_id** | **String** | Optional client-supplied ID echoed verbatim in the response. Use to correlate/sync with your system. Accepted charset: [A-Za-z0-9 _/:.-], max 60 chars. Not idempotency. | [optional] |
 
 ### Return type
 

data/docs/WhatsappApi.md

@@ -33,7 +33,8 @@ date_format = 'date_format_example' # String |
 opts = {
   gap_minutes: 8.14, # Float | 
   min_messages: 56, # Integer | 
-  chars_per_minute: 8.14, # Float | 
+  voice_note: true, # Boolean | Transcrever mensagens de voz/video gravadas na hora?
+  expression: true, # Boolean | Descrever expressoes visuais (figurinha/GIF/imagem)?
   client_reference_id: 'client_reference_id_example' # String | 
 }
 
@@ -75,7 +76,8 @@ end
 | **date_format** | **String** |  |  |
 | **gap_minutes** | **Float** |  | [optional][default to 720] |
 | **min_messages** | **Integer** |  | [optional][default to 2] |
-| **chars_per_minute** | **Float** |  | [optional][default to 800] |
+| **voice_note** | **Boolean** | Transcrever mensagens de voz/video gravadas na hora? | [optional][default to true] |
+| **expression** | **Boolean** | Descrever expressoes visuais (figurinha/GIF/imagem)? | [optional][default to true] |
 | **client_reference_id** | **String** |  | [optional] |
 
 ### Return type

data/docs/WhatsappMeta.md

@@ -13,13 +13,14 @@
 | **_end** | **String** | Window end (ISO) |  |
 | **gap_minutes** | **Float** | Gap used to split conversations |  |
 | **min_messages** | **Integer** | Minimum messages per conversation |  |
-| **chars_per_minute** | **Float** | Chars per minute used to estimate duration |  |
 | **turns** | **Integer** | Total parsed turns |  |
 | **system_lines** | **Integer** | System lines ignored |  |
 | **conversations_total** | **Integer** | Conversations before window filter |  |
 | **conversations_in_window** | **Integer** | Conversations overlapping the window |  |
 | **monologues_dropped** | **Integer** | Single-speaker conversations dropped |  |
 | **conversations_selected** | **Integer** | Final conversations returned |  |
+| **voice_note** | **Boolean** | Voice/video medias were transcribed |  |
+| **expression** | **Boolean** | Visual expressions were described |  |
 
 ## Example
 
@@ -36,13 +37,14 @@ instance = FalaAI::WhatsappMeta.new(
   _end: null,
   gap_minutes: null,
   min_messages: null,
-  chars_per_minute: null,
   turns: null,
   system_lines: null,
   conversations_total: null,
   conversations_in_window: null,
   monologues_dropped: null,
-  conversations_selected: null
+  conversations_selected: null,
+  voice_note: null,
+  expression: null
 )
 ```
 

data/docs/WhatsappUsage.md

@@ -6,7 +6,12 @@
 | ---- | ---- | ----------- | ----- |
 | **conversations** | **Integer** | Number of conversations returned |  |
 | **characters** | **Integer** | Total characters across conversations |  |
-| **credits_consumed** | **Integer** | Credits consumed (1 per conversation) |  |
+| **expressions** | **Integer** | Expression tokens across conversations (sticker/emoji/gif) |  |
+| **voice_notes** | **Integer** | Voice notes (PTT) resolved to text |  |
+| **conversation_credits** | **Integer** | Credits for conversations (2 each) |  |
+| **expression_credits** | **Float** | Credits for expressions (0.2 each) |  |
+| **audio_credits** | **Float** | Credits for audio duration (0.2 per second) |  |
+| **credits_consumed** | **Integer** | Total credits = ceil(conversation_credits + expression_credits + audio_credits) |  |
 | **processing_ms** | **Integer** | Total processing time in milliseconds |  |
 
 ## Example
@@ -17,6 +22,11 @@ require 'falaai-api'
 instance = FalaAI::WhatsappUsage.new(
   conversations: null,
   characters: null,
+  expressions: null,
+  voice_notes: null,
+  conversation_credits: null,
+  expression_credits: null,
+  audio_credits: null,
   credits_consumed: null,
   processing_ms: null
 )