Lexicost
AI voice agents / Vendors

Synthflow AI

In its own words: “Synthflow is an enterprise AI agent platform for customer conversations.”

6 of 11 key facts published, pages read .

Enterprise contact centre. Sales-led products for large contact centres. Most quote prices only on request.

What Synthflow AI publishes, in short. Platform fee per minute: . What that fee includes: (Enterprise). Monthly plan fee: . How the AI model is charged: . Every figure below opens Synthflow AI's own words and the date we read them.

Work out a cost with these figures: billed minutes: how vendors round a call, concurrency limits for AI voice agents, cost per call: AI agent or a person, aI voice agent cost per minute, openAI Realtime API cost per minute. Compare with other vendors in the side by side table.

Price

Platform fee per minuteEnterprise
What that fee includesEnterprise
Monthly plan feeEnterprise
Monthly plan feePro (Freshworks-connected workspaces)
Minimum commitmentEnterprise
How the AI model is charged
US phone minutes
PlanEnterprise
PlanPro (Freshworks-connected workspaces)
PlanPro, Growth, Agency (legacy)
Minutes includedPro (Freshworks-connected workspaces)
Price per minute beyond the planPro (Freshworks-connected workspaces)
Other phone rates
Phone number, per month

Billing rules

How a minute is billed
Billing rule
Billing rule
Billing rule
Billing rule
Billing rule
Billing ruleGTC Direct Use (US), customers with an Order Form
Billing ruleGTC Direct Use (US), customers with an Order Form
Billing rulesubscriptions managed in the Billing Portal

Capacity

Calls at once, includedEnterprise
Calls at once, includedPro (Freshworks-connected workspaces)
Each extra call at onceEnterprise
Each extra call at oncePro (Freshworks-connected workspaces)
When you exceed the limit
Other limits
Other limits
Other limits

Getting started

Free credit or trialEnterprise
Free credit or trialPro (Freshworks-connected workspaces)
Sign up without sales
Sign up without salesPro (Freshworks-connected workspaces)

Compliance

HIPAA agreement (BAA)
SOC 2
Where data is processed
Where data is processed
Other compliance statements
Other compliance statements
Other compliance statements
Other compliance statements

Product

Latency the vendor statesGTC Direct Use (US), customers with an Order Form
Latency the vendor states
Latency the vendor states
Latency the vendor states
Bring your own model, voice or carrierEnterprise
Bring your own model, voice or carrier
Bring your own model, voice or carrierEnterprise
White label for agenciesEnterprise
White label for agencies
Uptime commitmentEnterprise
Uptime commitmentSelf-serve subscribers (Subscriber Terms)
Uptime commitment
Support by plan
Support by planGTC Direct Use (US), Standard Support
Support by planGTC Direct Use (US), Premium Support
Turn-taking setting
Turn-taking setting
Turn-taking setting
Turn-taking setting
Turn-taking setting
Turn-taking setting
Turn-taking setting
How the vendor defines a resolved or contained call
How the vendor defines a resolved or contained call

Where Synthflow AI's own pages say different things

Both statements are quoted with their pages. We do not pick one. Ask the vendor which applies before you rely on either.

What Synthflow AI's public pages do not settle

  • .
  • .
  • .
  • .

What changed

  • 15 Sep 2026. In-product reselling removed: custom pricing plans and the Stripe integration no longer bill subaccounts (was Agencies could create pricing plans in Synthflow, connect Stripe and bill subaccounts automatically, now Partners bill subaccounts in their own systems and cap usage with Limits & Capabilities) source
  • 25 Aug 2026. Existing agents migrated automatically; GPT and other third-party models now require an Enterprise plan, and Synthflow LLM is the only model for non-enterprise customers source

Pages we read

Lexicost has no commercial relationship with any vendor named on this site. Spotted an error? See corrections.

Source for every vendor figure on this page (72)
  1. Synthflow AI, platform fee per minute: No per-minute rate; annual agreement (Enterprise. annual agreement).Synthflow is sold as an annual enterprise agreement, scoped to your call volume, phone system, integrations, security needs and launch support. It has no free plan or trial and no rate by the minute. synthflow.ai/pricing Read 14 Nov 2026.
  2. Synthflow AI, what that fee includes: Not stated for speech, AI model or voice (Enterprise. annual agreement. the agreement covers call concurrency and routing, telephony, integrations, security and procurement, support terms and, where included, launch support).The agreement covers call concurrency and routing, telephony, integrations, security and procurement (MSA and DPA support, an enterprise security review), support terms and, where included, launch support. synthflow.ai/facts Read 14 Nov 2026.
  3. Synthflow AI, monthly plan fee: No list price; priced after a scoping call (Enterprise. annual agreement).Synthflow is sold as an annual enterprise agreement, scoped to your call volume, phone system, integrations, security needs and launch support. It publishes no list price: its team prices each deployment after a scoping call. synthflow.ai/facts Read 14 Nov 2026.
  4. Synthflow AI, monthly plan fee: Pro, $400 per month (Freshworks users only) (Pro (Freshworks-connected workspaces). monthly. after a 14-day free trial; 2,000 minutes included each month).The Pro plan continues at $400 USD per month. docs.synthflow.ai/freshworks Read 14 Nov 2026.
  5. Synthflow AI, free credit or trial: No free plan or trial (Enterprise. annual agreement).No. Synthflow has no free plan or trial and no rate by the minute. synthflow.ai/pricing Read 14 Nov 2026.
  6. Synthflow AI, free credit or trial: 14-day free trial (Freshworks users only) (Pro (Freshworks-connected workspaces). the Pro plan continues at $400 per month after the trial).Freshworks-connected workspaces can subscribe to the Pro plan, which starts with a 14-day free trial. docs.synthflow.ai/freshworks Read 14 Nov 2026.
  7. Synthflow AI, sign up without sales: Enterprise contract for every new deployment.Synthflow is built for mid-market and enterprise teams deploying AI voice agents for customer calls, and every new deployment starts on an Enterprise contract. Also: Synthflow support has two paths: one for self-serve plans and one for Enterprise. Each has its own ticket form, so start with your plan. docs.synthflow.ai/getting-support Also: Version 1.0 | Last updated May 13, 2026 | Applicable to: Self-serve subscribers via synthflow.ai docs.synthflow.ai/terms-conditions Also: In some cases, new accounts may be reviewed before full workspace access is enabled. While your account is being reviewed, you may see a holding page confirming that your access request has been received. docs.synthflow.ai/account-access-login synthflow.ai/pricing Read 14 Nov 2026.
  8. Synthflow AI, sign up without sales: Yes for Freshworks users, via its Marketplace (Pro (Freshworks-connected workspaces). account created through the Freshworks Marketplace installation; existing Synthflow accounts cannot be migrated).Freshworks-connected workspaces can subscribe to the Pro plan, which starts with a 14-day free trial. For more capacity, custom telephony, or a security review, contact sales to scope an Enterprise agreement instead. docs.synthflow.ai/freshworks Read 14 Nov 2026.
  9. Synthflow AI, how the AI model is charged: Not publicly listed. Read 14 Nov 2026.
  10. Synthflow AI, uS phone minutes: Not publicly listed. Read 14 Nov 2026.
  11. Synthflow AI, how a minute is billed: Per second, partial seconds rounded up (measured from when the caller and the AI agent connect until the call ends).Phone calls are measured per second, starting when the human and the AI agent connect and ending when the call terminates. Partial seconds are rounded up to the nearest second. docs.synthflow.ai/usage Read 14 Nov 2026.
  12. Synthflow AI, calls at once, included: Not publicly listed (Enterprise). Read 14 Nov 2026.
  13. Synthflow AI, calls at once, included: 25 calls at once (Freshworks Pro plan) (Pro (Freshworks-connected workspaces). monthly).Concurrent calls: 25 docs.synthflow.ai/freshworks Read 14 Nov 2026.
  14. Synthflow AI, each extra call at once: Not publicly listed (Enterprise). Read 14 Nov 2026.
  15. Synthflow AI, each extra call at once: None on Freshworks Pro; Enterprise via sales (Pro (Freshworks-connected workspaces)).Not on the Pro plan. Contact sales to scope an Enterprise agreement with capacity matched to your call volume. docs.synthflow.ai/freshworks Read 14 Nov 2026.
  16. Synthflow AI, hIPAA agreement (BAA): Yes, for customers using the HIPAA Setting (customers using the HIPAA Setting to process protected health information; executed at contract signing).The BAA is executed between Synthflow and Customer at contract signing for customers using the HIPAA Setting to process Protected Health Information (PHI). docs.synthflow.ai/baa Read 14 Nov 2026.
  17. Synthflow AI, plan: Enterprise: annual agreement scoped to call volume, phone system, integrations, security and launch support (Enterprise. annual agreement).Synthflow is sold as an annual enterprise agreement, scoped to your call volume, phone system, integrations, security needs and launch support. synthflow.ai/pricing Read 14 Nov 2026.
  18. Synthflow AI, plan: Pro plan for Freshworks-connected workspaces, starting with a 14-day free trial (Pro (Freshworks-connected workspaces). monthly).Freshworks-connected workspaces can subscribe to the Pro plan, which starts with a 14-day free trial. For more capacity, custom telephony, or a security review, contact sales to scope an Enterprise agreement instead. docs.synthflow.ai/freshworks Read 14 Nov 2026.
  19. Synthflow AI, plan: Earlier Pro, Growth and Agency plans no longer sold (Pro, Growth, Agency (legacy)).Legacy workspaces are on the earlier Pro, Growth, and Agency plans, which came with a fixed concurrency limit and a default CAPS limit. These plans are no longer sold, and existing workspaces keep their limits. docs.synthflow.ai/concurrency Read 14 Nov 2026.
  20. Synthflow AI, minutes included: 2,000 minutes per month (Pro (Freshworks-connected workspaces). monthly).Included minutes: 2,000 per month docs.synthflow.ai/freshworks Read 14 Nov 2026.
  21. Synthflow AI, price per minute beyond the plan: $0.13 per minute beyond 2,000 included minutes (Pro (Freshworks-connected workspaces). monthly. minutes beyond the 2,000 included each month, on top of the monthly fee).Minutes beyond the 2,000 included each month are billed at $0.13 USD per minute on top of the monthly fee. docs.synthflow.ai/freshworks Read 14 Nov 2026.
  22. Synthflow AI, minimum commitment: Not publicly listed (Enterprise). Read 14 Nov 2026.
  23. Synthflow AI, phone number, per month: $1.50 per month for a number bought through Synthflow (U.S., Canadian and Australian numbers bought through Synthflow, described as for quick testing).Synthflow offers U.S., Canadian, and Australian numbers for quick testing directly through the platform. Phone numbers cost $1.50 USD per month. docs.synthflow.ai/phone-numbers Read 14 Nov 2026.
  24. Synthflow AI, other phone rates: International calling priced per agreement.International calling can vary by destination, carrier path, and telephony setup. Review your agreement or contact your account team for details. docs.synthflow.ai/usage Read 14 Nov 2026.
  25. Synthflow AI, billing rule: Seconds are added up across calls; each whole minute is billed.Rather than rounding each individual call, Synthflow aggregates seconds across calls. Each whole minute that accumulates is reported as a billing event on your invoice, so usage reflects your actual call activity. docs.synthflow.ai/usage Read 14 Nov 2026.
  26. Synthflow AI, billing rule: No-answer and hang-up-on-voicemail calls count as 5 seconds; failed and canceled calls are not billed.Calls with a no-answer or hangup_on_voicemail status are recorded as a flat 5 seconds of usage. Calls with a failed or user-canceled status are not billed. docs.synthflow.ai/usage Read 14 Nov 2026.
  27. Synthflow AI, billing rule: Billing stops when a transfer completes.If a transfer completes successfully, Synthflow stops measuring billable AI usage at the point of transfer. Usage after the transfer is not included. docs.synthflow.ai/usage Read 14 Nov 2026.
  28. Synthflow AI, billing rule: Chat: every 5 AI-generated messages count as 1 usage minute (live chat through API, widget, SMS and similar channels).Live chat through API, widget, SMS, and similar channels is converted into billable usage minutes: every 5 AI-generated messages count as 1 usage minute. docs.synthflow.ai/usage Read 14 Nov 2026.
  29. Synthflow AI, billing rule: Billing thresholds trigger an invoice; they are not spending caps (accounts with threshold invoicing).Billing thresholds are triggers, not spending caps. When unbilled usage reaches the configured threshold for your account, an invoice can be generated for the usage accrued up to that moment. docs.synthflow.ai/usage Read 14 Nov 2026.
  30. Synthflow AI, billing rule: Fees paid are non-refundable and ordered usage volume cannot be decreased, unless the order form says otherwise (GTC Direct Use (US), customers with an Order Form).Except as otherwise specified herein or in the applicable Order Form Fees paid are non-refundable and Usage Volume ordered cannot be decreased. docs.synthflow.ai/gtc-direct-us Read 14 Nov 2026.
  31. Synthflow AI, billing rule: Invoiced in advance at the start of each billing period, unless the order form says otherwise (GTC Direct Use (US), customers with an Order Form).If not agreed upon otherwise in the Order Form, Synthflow shall invoice the Customer in advance upon the commencement of each Billing Period. docs.synthflow.ai/gtc-direct-us Read 14 Nov 2026.
  32. Synthflow AI, billing rule: No refunds for partial billing periods after cancelling a subscription (subscriptions managed in the Billing Portal).Refunds are not issued for partial billing periods. docs.synthflow.ai/manage-billing Read 14 Nov 2026.
  33. Synthflow AI, when you exceed the limit: Extra calls get an HTTP 429 error or wait briefly in a queue, depending on account setup.When you reach a limit, extra calls either return an HTTP 429 error or wait briefly in a queue, depending on your account setup. docs.synthflow.ai/concurrency Read 14 Nov 2026.
  34. Synthflow AI, other limits: Outbound call starts per second (CAPS) limited per account, across API calls, campaigns and warm transfers.CAPS (call attempts per second): How many outbound calls your account can start per second, including calls from the Make a Phone Call API, campaigns, and warm transfers. It applies to every phone number type. docs.synthflow.ai/concurrency Read 14 Nov 2026.
  35. Synthflow AI, other limits: Each warm transfer attempt counts against CAPS.Every warm transfer attempt starts a new outbound call to the transfer destination, so it counts against CAPS. docs.synthflow.ai/concurrency Read 14 Nov 2026.
  36. Synthflow AI, other limits: Subaccounts share the main account's limits (agency or white-label subaccounts).Subaccounts do not get their own limits. All subaccounts share the main account’s total. docs.synthflow.ai/concurrency Read 14 Nov 2026.
  37. Synthflow AI, sOC 2: SOC 2 Type 2 report, requested through the Trust Center.Synthflow's SOC 2 Type 2 report covers how its controls operated over a defined audit period. Request it through the Trust Center. synthflow.ai/security Read 14 Nov 2026.
  38. Synthflow AI, where data is processed: US or EU data region, chosen when a workspace is created.You choose a US or EU data region when you create a workspace, which determines where data is stored and processed. synthflow.ai/security Read 14 Nov 2026.
  39. Synthflow AI, where data is processed: Legacy Global cluster: data stored in the EU, live calls processed in the US or EU (workspaces created before the US and EU clusters existed).Global is the default for workspaces created before dedicated US and EU clusters were available. Data for Global workspaces is stored in the EU, but live call processing can run in the United States or the European Union depending on traffic and routing. docs.synthflow.ai/customer-region Read 14 Nov 2026.
  40. Synthflow AI, other compliance statements: Trust Center list, as given on the pricing page: SOC 2 Type 2, ISO/IEC 27001, GDPR, HIPAA (self-attested), PCI DSS v4.0.1; ISO/IEC 42001, DORA and the EU AI Act shown as in progress.Synthflow's Trust Center lists SOC 2 Type 2, ISO/IEC 27001, GDPR, HIPAA (self-attested) and PCI DSS v4.0.1, and shows ISO/IEC 42001:2023, DORA and the EU AI Act as in progress. synthflow.ai/pricing Read 14 Nov 2026.
  41. Synthflow AI, other compliance statements: Security page: ISO/IEC 42001 and the EU AI Act listed as AI governance aligned, not as certifications.Synthflow's Trust Center lists ISO/IEC 42001:2023 and the EU AI Act, and states that its AI governance program is aligned with both. Neither is listed as a certification. synthflow.ai/security Read 14 Nov 2026.
  42. Synthflow AI, other compliance statements: Customer conversation data and customer knowledge not used to train Synthflow's models.Synthflow does not use customer conversation data or customer-provided knowledge to train its AI models. synthflow.ai/security Read 14 Nov 2026.
  43. Synthflow AI, other compliance statements: HIPAA Setting: protected health information not logged, recorded or stored persistently (HIPAA Setting switched on (off by default)).With the HIPAA Setting on, Synthflow does not log, record, or persistently store protected health information. synthflow.ai/security Read 14 Nov 2026.
  44. Synthflow AI, latency the vendor states: SLA test: 90% of bot responses within 2,800 ms (GTC Direct Use (US), customers with an Order Form. synthetic end-to-end calls, reference prompt under 10,000 tokens, no RAG or external actions, measured from end of caller speech to agent speech (STT + LLMs + TTS + telephony), tested every 3 minutes).P90 ≤ 2,800 ms bot response time. docs.synthflow.ai/gtc-direct-us Read 14 Nov 2026.
  45. Synthflow AI, latency the vendor states: Synthflow TTS: under 200 ms to first audio (Synthflow TTS voice model, time to first audio).Low latency: Optimized for real-time conversational use cases with sub-200ms time-to-first-audio. docs.synthflow.ai/voice-configuration Read 14 Nov 2026.
  46. Synthflow AI, latency the vendor states: Target round-trip time under 100 ms for traffic within a region (telephony traffic that stays within a region; depends on access network, peering and carrier path).For traffic that stays within a region, we target a round-trip time under 100 ms and a MOS above 4.2. docs.synthflow.ai/telephony Read 14 Nov 2026.
  47. Synthflow AI, latency the vendor states: Blog table: average latency 500 to 700 ms; what is measured not stated (comparison table in a Synthflow blog post dated July 3, 2025; the post does not say what the latency covers or how it was measured).Average latency: 500 to 700ms Also: Synthflow offers latency under 500ms, dynamic emotion mapping, and multilingual voice cloning. synthflow.ai/blog/parloa-review synthflow.ai/blog/how-can-i-use-ai-to-reduce-the-number-of-call-center-agents Read 14 Nov 2026.
  48. Synthflow AI, bring your own model, voice or carrier: Own carrier or PBX over SIP, Enterprise plan only (Enterprise).Enterprise Feature: SIP/PBX integration requires an Enterprise plan subscription. Contact our sales team to enable this feature. docs.synthflow.ai/telephony Read 14 Nov 2026.
  49. Synthflow AI, bring your own model, voice or carrier: Bring your own Twilio numbers.Bring your Twilio numbers: Keep numbers in your Twilio account and route them through Synthflow. docs.synthflow.ai/phone-numbers Read 14 Nov 2026.
  50. Synthflow AI, bring your own model, voice or carrier: Only an OpenAI API key, on Enterprise plans; no Claude or custom LLM endpoint (Enterprise).No. The AI Model selector does not support Claude or custom LLM endpoints. The only bring-your-own option is an OpenAI API key through BYOK, which runs GPT models and so applies to Enterprise plans. docs.synthflow.ai/general-configuration Read 14 Nov 2026.
  51. Synthflow AI, white label for agencies: Yes, in the annual enterprise agreement (Enterprise. annual agreement).Yes, with white-label in the annual enterprise agreement. You give each client its own workspace under your brand and domain, and see every client in one analytics dashboard. synthflow.ai/pricing Read 14 Nov 2026.
  52. Synthflow AI, white label for agencies: In-product billing of subaccounts removed on September 15, 2026 (agency and white-label subaccounts).In-product reselling let you create pricing plans in Synthflow, connect Stripe, and bill subaccounts automatically. That capability was removed on September 15, 2026. docs.synthflow.ai/reselling-migration-guide Read 14 Nov 2026.
  53. Synthflow AI, uptime commitment: Availability and support terms set in each Enterprise agreement (Enterprise).Contracted support, availability, incident, and governance expectations are documented in your agreement instead of relying on public default claims. docs.synthflow.ai/enterprise Read 14 Nov 2026.
  54. Synthflow AI, uptime commitment: Self-serve: commercially reasonable efforts to be available 24/7, with exceptions (Self-serve subscribers (Subscriber Terms)).Synthflow shall use commercially reasonable efforts to make the Software available 24 hours a day, 7 days a week, except for: (i) planned downtime (of which Synthflow shall give notice in advance); and (ii) any unavailability caused by circumstances beyond Synthflow’s reasonable control, including, for example, internet service provider failure or delay, non-Synthflow applications, or denial of service attack. docs.synthflow.ai/terms-conditions Read 14 Nov 2026.
  55. Synthflow AI, uptime commitment: 99.99% platform uptime (stated figure).Platform uptime: 99.99% synthflow.ai/facts Read 14 Nov 2026.
  56. Synthflow AI, support by plan: Separate support paths for self-serve plans and Enterprise.Synthflow support has two paths: one for self-serve plans and one for Enterprise. Each has its own ticket form, so start with your plan. docs.synthflow.ai/getting-support Read 14 Nov 2026.
  57. Synthflow AI, support by plan: Standard Support: critical issues acknowledged within 60 minutes (GTC Direct Use (US), Standard Support. priority 0, critical or service down).Critical / Service Down: Within 60 minutes docs.synthflow.ai/gtc-direct-us Read 14 Nov 2026.
  58. Synthflow AI, support by plan: Premium Support: critical issues acknowledged within 30 minutes (GTC Direct Use (US), Premium Support. priority 0, critical or service down).Critical / Service Down: Within 30 minutes docs.synthflow.ai/gtc-direct-us Read 14 Nov 2026.
  59. Synthflow AI, where the vendor's own pages disagree: The concurrency docs say the earlier Pro, Growth and Agency plans are no longer sold, while the Freshworks docs say Freshworks-connected workspaces can subscribe to a Pro plan..These plans are no longer sold, and existing workspaces keep their limits. Also: Freshworks-connected workspaces can subscribe to the Pro plan, which starts with a 14-day free trial. docs.synthflow.ai/freshworks docs.synthflow.ai/concurrency Read 14 Nov 2026.
  60. Synthflow AI, where the vendor's own pages disagree: The pricing page says every new deployment starts on an Enterprise contract, while the support docs and the Subscriber Terms describe self-serve plans and subscribers via synthflow.ai..Synthflow is built for mid-market and enterprise teams deploying AI voice agents for customer calls, and every new deployment starts on an Enterprise contract. Also: Synthflow support has two paths: one for self-serve plans and one for Enterprise. Each has its own ticket form, so start with your plan. docs.synthflow.ai/getting-support Also: Version 1.0 | Last updated May 13, 2026 | Applicable to: Self-serve subscribers via synthflow.ai docs.synthflow.ai/terms-conditions synthflow.ai/pricing Read 14 Nov 2026.
  61. Synthflow AI, where the vendor's own pages disagree: Two Synthflow blog posts give different latency figures for Synthflow, and neither says what it measures: an average of 500 to 700 ms (July 3, 2025) and under 500 ms (May 21, 2025)..Average latency: 500 to 700ms Also: Synthflow offers latency under 500ms, dynamic emotion mapping, and multilingual voice cloning. synthflow.ai/blog/parloa-review synthflow.ai/blog/how-can-i-use-ai-to-reduce-the-number-of-call-center-agents Read 14 Nov 2026.
  62. Synthflow AI, where the vendor's own pages disagree: The agent editor docs name the setting Interruption Sensitivity, with an Off choice, while the API reference has a field min_words_to_interrupt (0 to 5) with the same description; neither page says they are one setting..Interruption Sensitivity controls how easily the caller can interrupt the agent mid-sentence. The value represents how many words the caller must speak before the agent stops and listens. Setting it to Off disables interruptions entirely; lower values make the agent yield quickly, while higher values let it finish longer phrases before pausing. Also: You can set how many words the human must speak before the AI will stop speaking and listening. docs.synthflow.ai/api-reference/platform-api/agents/create-assistant docs.synthflow.ai/additional-settings Read 14 Nov 2026.
  63. Synthflow AI, where the vendor's own pages disagree: The agent editor docs describe Pause Before Speaking, while the API reference has two fields, initial_pause_seconds and ring_pause_seconds, and neither page says which one it is..Pause Before Speaking adds a delay (in seconds) before the agent begins talking at the start of a call. Also: Duration before the agent starts speaking. docs.synthflow.ai/api-reference/platform-api/agents/create-assistant Also: Ring time before the agent answers the call. docs.synthflow.ai/api-reference/platform-api/agents/create-assistant docs.synthflow.ai/additional-settings Read 14 Nov 2026.
  64. Synthflow AI, turn-taking setting: Patience Level: how long the agent waits after the caller finishes before it responds: Low, Medium or High (Values: low, medium, high (API); agent editor, Voice section).Patience Level determines how long the agent waits after the caller finishes speaking before it begins its response. Low makes the agent respond almost immediately, Medium adds a natural pause, and High gives the caller extra time to continue. docs.synthflow.ai/voice-configuration Read 14 Nov 2026.
  65. Synthflow AI, turn-taking setting: Interruption Sensitivity: words the caller must speak before the agent stops and listens; Off disables it (Agent editor, Additional Settings).Interruption Sensitivity controls how easily the caller can interrupt the agent mid-sentence. The value represents how many words the caller must speak before the agent stops and listens. Setting it to Off disables interruptions entirely; lower values make the agent yield quickly, while higher values let it finish longer phrases before pausing. Also: You can set how many words the human must speak before the AI will stop speaking and listening. docs.synthflow.ai/api-reference/platform-api/agents/create-assistant docs.synthflow.ai/additional-settings Read 14 Nov 2026.
  66. Synthflow AI, turn-taking setting: Fade Out at Interruption: how many frames the agent's voice takes to fade out once an interruption is detected (Frames of 20 ms each (API reference: 5 frames is a 100 ms fade)).Fade Out at Interruption sets how many frames the agent’s voice takes to fade out once an interruption is detected. docs.synthflow.ai/additional-settings Read 14 Nov 2026.
  67. Synthflow AI, turn-taking setting: Pause Before Speaking: a delay in seconds before the agent begins talking at the start of a call (Agent editor, Additional Settings; seconds).Pause Before Speaking adds a delay (in seconds) before the agent begins talking at the start of a call. Also: Duration before the agent starts speaking. docs.synthflow.ai/api-reference/platform-api/agents/create-assistant Also: Ring time before the agent answers the call. docs.synthflow.ai/api-reference/platform-api/agents/create-assistant docs.synthflow.ai/additional-settings Read 14 Nov 2026.
  68. Synthflow AI, turn-taking setting: greeting_message_mode: human waits for the caller to speak first; agent_static or agent_dynamic greet on connect (Allowed values: human, agent_static, agent_dynamic; voicemail messages need human).Controls how the agent delivers the initial greeting. human: the agent waits for the human to speak first. agent_static: the agent speaks greeting_message once the call is established. agent_dynamic: the agent generates a greeting from the prompt and speaks it once the call is established. docs.synthflow.ai/api-reference/platform-api/agents/create-assistant Read 14 Nov 2026.
  69. Synthflow AI, turn-taking setting: How long to wait: with Human speaks first, seconds of silence (1 to 10) before the agent starts the conversation (1 to 10 seconds; only when the greeting is set to Human speaks first).The How long to wait slider sets a fallback timeout (1 to 10 seconds). If the caller stays silent past that point, the agent steps in and starts the conversation on its own. Three seconds is a balanced default. docs.synthflow.ai/greeting-message-behavior Read 14 Nov 2026.
  70. Synthflow AI, turn-taking setting: Idle reminders: the agent sends a reminder message after the caller has been silent for a set period (Message and Idle Duration threshold are set by the customer; agent editor, Call Configuration, Advanced settings).When enabled, the agent sends a reminder message after the caller has been silent for a set period. You can customize the Message the agent says (for example, “I’m still here. Do you have any questions?”) and the Idle Duration threshold that triggers it. docs.synthflow.ai/call-configuration Read 14 Nov 2026.
  71. Synthflow AI, how the vendor defines a resolved or contained call: The customer writes the criteria for what makes a successful call (custom evaluations) (Custom evaluation actions attached to an agent).Custom evaluations allow you to define and track your own criteria for what makes a successful call or conversation. docs.synthflow.ai/configure-custom-evaluations Read 14 Nov 2026.
  72. Synthflow AI, how the vendor defines a resolved or contained call: Not publicly listed. Read 14 Nov 2026.