hume 0.11.4 → 0.11.5-beta.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (109) hide show
  1. package/.mock/definition/empathic-voice/__package__.yml +186 -61
  2. package/.mock/definition/empathic-voice/chat.yml +1 -8
  3. package/.mock/definition/empathic-voice/configs.yml +7 -1
  4. package/.mock/definition/tts/__package__.yml +17 -14
  5. package/Client.js +2 -2
  6. package/api/resources/empathicVoice/resources/chat/types/SubscribeEvent.d.ts +1 -1
  7. package/api/resources/empathicVoice/resources/configs/client/requests/PostedConfig.d.ts +1 -0
  8. package/api/resources/empathicVoice/types/AssistantProsody.d.ts +21 -0
  9. package/api/resources/empathicVoice/types/AssistantProsody.js +5 -0
  10. package/api/resources/empathicVoice/types/PostedNudgeSpec.d.ts +2 -2
  11. package/api/resources/empathicVoice/types/ReturnConfig.d.ts +19 -19
  12. package/api/resources/empathicVoice/types/ReturnPrompt.d.ts +12 -12
  13. package/api/resources/empathicVoice/types/ReturnSignedUri.d.ts +12 -0
  14. package/api/resources/empathicVoice/types/ReturnSignedUri.js +5 -0
  15. package/api/resources/empathicVoice/types/ReturnVoiceOctaveCustom.d.ts +31 -0
  16. package/api/resources/empathicVoice/types/ReturnVoiceOctaveCustom.js +5 -0
  17. package/api/resources/empathicVoice/types/ReturnVoiceOctaveShared.d.ts +15 -0
  18. package/api/resources/empathicVoice/types/ReturnVoiceOctaveShared.js +5 -0
  19. package/api/resources/empathicVoice/types/ReturnVoiceSpec.d.ts +8 -0
  20. package/api/resources/empathicVoice/types/ReturnVoiceSpec.js +5 -0
  21. package/api/resources/empathicVoice/types/ReturnVoiceSpecEvi2.d.ts +14 -0
  22. package/api/resources/empathicVoice/types/ReturnVoiceSpecEvi2.js +5 -0
  23. package/api/resources/empathicVoice/types/ReturnVoiceSpecOctavecustom.d.ts +14 -0
  24. package/api/resources/empathicVoice/types/ReturnVoiceSpecOctavecustom.js +5 -0
  25. package/api/resources/empathicVoice/types/ReturnVoiceSpecOctaveshared.d.ts +14 -0
  26. package/api/resources/empathicVoice/types/ReturnVoiceSpecOctaveshared.js +5 -0
  27. package/api/resources/empathicVoice/types/index.d.ts +9 -1
  28. package/api/resources/empathicVoice/types/index.js +9 -1
  29. package/api/resources/tts/types/PostedTts.d.ts +2 -3
  30. package/api/resources/tts/types/PostedUtterance.d.ts +5 -3
  31. package/dist/Client.js +2 -2
  32. package/dist/api/resources/empathicVoice/resources/chat/types/SubscribeEvent.d.ts +1 -1
  33. package/dist/api/resources/empathicVoice/resources/configs/client/requests/PostedConfig.d.ts +1 -0
  34. package/dist/api/resources/empathicVoice/types/AssistantProsody.d.ts +21 -0
  35. package/dist/api/resources/empathicVoice/types/AssistantProsody.js +5 -0
  36. package/dist/api/resources/empathicVoice/types/PostedNudgeSpec.d.ts +2 -2
  37. package/dist/api/resources/empathicVoice/types/ReturnConfig.d.ts +19 -19
  38. package/dist/api/resources/empathicVoice/types/ReturnPrompt.d.ts +12 -12
  39. package/dist/api/resources/empathicVoice/types/ReturnSignedUri.d.ts +12 -0
  40. package/dist/api/resources/empathicVoice/types/ReturnSignedUri.js +5 -0
  41. package/dist/api/resources/empathicVoice/types/ReturnVoiceOctaveCustom.d.ts +31 -0
  42. package/dist/api/resources/empathicVoice/types/ReturnVoiceOctaveCustom.js +5 -0
  43. package/dist/api/resources/empathicVoice/types/ReturnVoiceOctaveShared.d.ts +15 -0
  44. package/dist/api/resources/empathicVoice/types/ReturnVoiceOctaveShared.js +5 -0
  45. package/dist/api/resources/empathicVoice/types/ReturnVoiceSpec.d.ts +8 -0
  46. package/dist/api/resources/empathicVoice/types/ReturnVoiceSpec.js +5 -0
  47. package/dist/api/resources/empathicVoice/types/ReturnVoiceSpecEvi2.d.ts +14 -0
  48. package/dist/api/resources/empathicVoice/types/ReturnVoiceSpecEvi2.js +5 -0
  49. package/dist/api/resources/empathicVoice/types/ReturnVoiceSpecOctavecustom.d.ts +14 -0
  50. package/dist/api/resources/empathicVoice/types/ReturnVoiceSpecOctavecustom.js +5 -0
  51. package/dist/api/resources/empathicVoice/types/ReturnVoiceSpecOctaveshared.d.ts +14 -0
  52. package/dist/api/resources/empathicVoice/types/ReturnVoiceSpecOctaveshared.js +5 -0
  53. package/dist/api/resources/empathicVoice/types/index.d.ts +9 -1
  54. package/dist/api/resources/empathicVoice/types/index.js +9 -1
  55. package/dist/api/resources/tts/types/PostedTts.d.ts +2 -3
  56. package/dist/api/resources/tts/types/PostedUtterance.d.ts +5 -3
  57. package/dist/serialization/resources/empathicVoice/resources/chat/types/SubscribeEvent.d.ts +2 -1
  58. package/dist/serialization/resources/empathicVoice/resources/chat/types/SubscribeEvent.js +2 -0
  59. package/dist/serialization/resources/empathicVoice/types/AssistantProsody.d.ts +16 -0
  60. package/dist/serialization/resources/empathicVoice/types/AssistantProsody.js +47 -0
  61. package/dist/serialization/resources/empathicVoice/types/ReturnConfig.d.ts +15 -15
  62. package/dist/serialization/resources/empathicVoice/types/ReturnConfig.js +15 -15
  63. package/dist/serialization/resources/empathicVoice/types/ReturnPrompt.d.ts +3 -3
  64. package/dist/serialization/resources/empathicVoice/types/ReturnPrompt.js +3 -3
  65. package/dist/serialization/resources/empathicVoice/types/ReturnSignedUri.d.ts +15 -0
  66. package/dist/serialization/resources/empathicVoice/types/ReturnSignedUri.js +46 -0
  67. package/dist/serialization/resources/empathicVoice/types/ReturnVoiceOctaveCustom.d.ts +34 -0
  68. package/dist/serialization/resources/empathicVoice/types/ReturnVoiceOctaveCustom.js +65 -0
  69. package/dist/serialization/resources/empathicVoice/types/ReturnVoiceOctaveShared.d.ts +18 -0
  70. package/dist/serialization/resources/empathicVoice/types/ReturnVoiceOctaveShared.js +49 -0
  71. package/dist/serialization/resources/empathicVoice/types/ReturnVoiceSpec.d.ts +13 -0
  72. package/dist/serialization/resources/empathicVoice/types/ReturnVoiceSpec.js +48 -0
  73. package/dist/serialization/resources/empathicVoice/types/ReturnVoiceSpecEvi2.d.ts +15 -0
  74. package/dist/serialization/resources/empathicVoice/types/ReturnVoiceSpecEvi2.js +46 -0
  75. package/dist/serialization/resources/empathicVoice/types/ReturnVoiceSpecOctavecustom.d.ts +15 -0
  76. package/dist/serialization/resources/empathicVoice/types/ReturnVoiceSpecOctavecustom.js +46 -0
  77. package/dist/serialization/resources/empathicVoice/types/ReturnVoiceSpecOctaveshared.d.ts +15 -0
  78. package/dist/serialization/resources/empathicVoice/types/ReturnVoiceSpecOctaveshared.js +46 -0
  79. package/dist/serialization/resources/empathicVoice/types/index.d.ts +9 -1
  80. package/dist/serialization/resources/empathicVoice/types/index.js +9 -1
  81. package/dist/version.d.ts +1 -1
  82. package/dist/version.js +1 -1
  83. package/package.json +1 -1
  84. package/serialization/resources/empathicVoice/resources/chat/types/SubscribeEvent.d.ts +2 -1
  85. package/serialization/resources/empathicVoice/resources/chat/types/SubscribeEvent.js +2 -0
  86. package/serialization/resources/empathicVoice/types/AssistantProsody.d.ts +16 -0
  87. package/serialization/resources/empathicVoice/types/AssistantProsody.js +47 -0
  88. package/serialization/resources/empathicVoice/types/ReturnConfig.d.ts +15 -15
  89. package/serialization/resources/empathicVoice/types/ReturnConfig.js +15 -15
  90. package/serialization/resources/empathicVoice/types/ReturnPrompt.d.ts +3 -3
  91. package/serialization/resources/empathicVoice/types/ReturnPrompt.js +3 -3
  92. package/serialization/resources/empathicVoice/types/ReturnSignedUri.d.ts +15 -0
  93. package/serialization/resources/empathicVoice/types/ReturnSignedUri.js +46 -0
  94. package/serialization/resources/empathicVoice/types/ReturnVoiceOctaveCustom.d.ts +34 -0
  95. package/serialization/resources/empathicVoice/types/ReturnVoiceOctaveCustom.js +65 -0
  96. package/serialization/resources/empathicVoice/types/ReturnVoiceOctaveShared.d.ts +18 -0
  97. package/serialization/resources/empathicVoice/types/ReturnVoiceOctaveShared.js +49 -0
  98. package/serialization/resources/empathicVoice/types/ReturnVoiceSpec.d.ts +13 -0
  99. package/serialization/resources/empathicVoice/types/ReturnVoiceSpec.js +48 -0
  100. package/serialization/resources/empathicVoice/types/ReturnVoiceSpecEvi2.d.ts +15 -0
  101. package/serialization/resources/empathicVoice/types/ReturnVoiceSpecEvi2.js +46 -0
  102. package/serialization/resources/empathicVoice/types/ReturnVoiceSpecOctavecustom.d.ts +15 -0
  103. package/serialization/resources/empathicVoice/types/ReturnVoiceSpecOctavecustom.js +46 -0
  104. package/serialization/resources/empathicVoice/types/ReturnVoiceSpecOctaveshared.d.ts +15 -0
  105. package/serialization/resources/empathicVoice/types/ReturnVoiceSpecOctaveshared.js +46 -0
  106. package/serialization/resources/empathicVoice/types/index.d.ts +9 -1
  107. package/serialization/resources/empathicVoice/types/index.js +9 -1
  108. package/version.d.ts +1 -1
  109. package/version.js +1 -1
@@ -92,6 +92,31 @@ types:
92
92
  output.
93
93
  source:
94
94
  openapi: evi-asyncapi.json
95
+ AssistantProsody:
96
+ docs: When provided, the output is an Assistant Prosody message.
97
+ properties:
98
+ custom_session_id:
99
+ type: optional<string>
100
+ docs: >-
101
+ Used to manage conversational state, correlate frontend and backend
102
+ data, and persist conversations across EVI sessions.
103
+ id:
104
+ type: optional<string>
105
+ docs: Unique identifier for the segment.
106
+ models:
107
+ type: Inference
108
+ docs: Inference model results.
109
+ type:
110
+ type: literal<"assistant_prosody">
111
+ docs: >-
112
+ The type of message sent through the socket; for an Assistant Prosody
113
+ message, this must be `assistant_PROSODY`.
114
+
115
+
116
+ This message the expression measurement predictions of the assistant's
117
+ audio output.
118
+ source:
119
+ openapi: evi-asyncapi.json
95
120
  AudioConfiguration:
96
121
  properties:
97
122
  channels:
@@ -1340,22 +1365,30 @@ types:
1340
1365
  type: list<optional<ReturnPrompt>>
1341
1366
  source:
1342
1367
  openapi: evi-openapi.json
1343
- ReturnPromptVersionType:
1344
- enum:
1345
- - FIXED
1346
- - LATEST
1347
- docs: >-
1348
- Versioning method for a Prompt. Either `FIXED` for using a fixed version
1349
- number or `LATEST` for auto-updating to the latest version.
1350
- inline: true
1351
- source:
1352
- openapi: evi-openapi.json
1353
1368
  ReturnPrompt:
1354
1369
  docs: A Prompt associated with this Config.
1355
1370
  properties:
1371
+ name:
1372
+ type: string
1373
+ docs: Name applied to all versions of a particular Prompt.
1356
1374
  id:
1357
1375
  type: string
1358
1376
  docs: Identifier for a Prompt. Formatted as a UUID.
1377
+ text:
1378
+ type: string
1379
+ docs: >-
1380
+ Instructions used to shape EVI’s behavior, responses, and style.
1381
+
1382
+
1383
+ You can use the Prompt to define a specific goal or role for EVI,
1384
+ specifying how it should act or what it should focus on during the
1385
+ conversation. For example, EVI can be instructed to act as a customer
1386
+ support representative, a fitness coach, or a travel advisor, each
1387
+ with its own set of behaviors and response styles.
1388
+
1389
+
1390
+ For help writing a system prompt, see our [Prompting
1391
+ Guide](/docs/empathic-voice-interface-evi/guides/prompting).
1359
1392
  version:
1360
1393
  type: integer
1361
1394
  docs: >-
@@ -1376,12 +1409,6 @@ types:
1376
1409
  docs: >-
1377
1410
  Versioning method for a Prompt. Either `FIXED` for using a fixed
1378
1411
  version number or `LATEST` for auto-updating to the latest version.
1379
- version_description:
1380
- type: optional<string>
1381
- docs: An optional description of the Prompt version.
1382
- name:
1383
- type: string
1384
- docs: Name applied to all versions of a particular Prompt.
1385
1412
  created_on:
1386
1413
  type: long
1387
1414
  docs: >-
@@ -1392,21 +1419,9 @@ types:
1392
1419
  docs: >-
1393
1420
  Time at which the Prompt was last modified. Measured in seconds since
1394
1421
  the Unix epoch.
1395
- text:
1396
- type: string
1397
- docs: >-
1398
- Instructions used to shape EVI’s behavior, responses, and style.
1399
-
1400
-
1401
- You can use the Prompt to define a specific goal or role for EVI,
1402
- specifying how it should act or what it should focus on during the
1403
- conversation. For example, EVI can be instructed to act as a customer
1404
- support representative, a fitness coach, or a travel advisor, each
1405
- with its own set of behaviors and response styles.
1406
-
1407
-
1408
- For help writing a system prompt, see our [Prompting
1409
- Guide](/docs/empathic-voice-interface-evi/guides/prompting).
1422
+ version_description:
1423
+ type: optional<string>
1424
+ docs: An optional description of the Prompt version.
1410
1425
  source:
1411
1426
  openapi: evi-openapi.json
1412
1427
  ReturnPagedCustomVoices:
@@ -1798,6 +1813,9 @@ types:
1798
1813
  ReturnConfig:
1799
1814
  docs: A specific config version returned from the server
1800
1815
  properties:
1816
+ name:
1817
+ type: optional<string>
1818
+ docs: Name applied to all versions of a particular Config.
1801
1819
  id:
1802
1820
  type: optional<string>
1803
1821
  docs: Identifier for a Config. Formatted as a UUID.
@@ -1823,12 +1841,28 @@ types:
1823
1841
  for the latest enhanced version. For a detailed comparison of the two
1824
1842
  versions, refer to our
1825
1843
  [guide](/docs/empathic-voice-interface-evi/configuration/evi-version).
1826
- version_description:
1827
- type: optional<string>
1828
- docs: An optional description of the Config version.
1829
- name:
1830
- type: optional<string>
1831
- docs: Name applied to all versions of a particular Config.
1844
+ timeouts: optional<ReturnTimeoutSpecs>
1845
+ nudges: optional<ReturnNudgeSpec>
1846
+ ellm_model:
1847
+ type: optional<ReturnEllmModel>
1848
+ docs: >-
1849
+ The eLLM setup associated with this Config.
1850
+
1851
+
1852
+ Hume's eLLM (empathic Large Language Model) is a multimodal language
1853
+ model that takes into account both expression measures and language.
1854
+ The eLLM generates short, empathic language responses and guides
1855
+ text-to-speech (TTS) prosody.
1856
+ voice:
1857
+ type: optional<ReturnVoice>
1858
+ docs: A voice specification associated with this Config.
1859
+ prompt: optional<ReturnPrompt>
1860
+ tools:
1861
+ type: optional<list<optional<ReturnUserDefinedTool>>>
1862
+ docs: List of user-defined tools associated with this Config.
1863
+ webhooks:
1864
+ type: optional<list<optional<ReturnWebhookSpec>>>
1865
+ docs: Map of webhooks associated with this config.
1832
1866
  created_on:
1833
1867
  type: optional<long>
1834
1868
  docs: >-
@@ -1839,10 +1873,6 @@ types:
1839
1873
  docs: >-
1840
1874
  Time at which the Config was last modified. Measured in seconds since
1841
1875
  the Unix epoch.
1842
- prompt: optional<ReturnPrompt>
1843
- voice:
1844
- type: optional<ReturnVoice>
1845
- docs: A voice specification associated with this Config.
1846
1876
  language_model:
1847
1877
  type: optional<ReturnLanguageModel>
1848
1878
  docs: >-
@@ -1852,28 +1882,13 @@ types:
1852
1882
  This model is used to generate longer, more detailed responses from
1853
1883
  EVI. Choosing an appropriate supplemental language model for your use
1854
1884
  case is crucial for generating fast, high-quality responses from EVI.
1855
- ellm_model:
1856
- type: optional<ReturnEllmModel>
1857
- docs: >-
1858
- The eLLM setup associated with this Config.
1859
-
1860
-
1861
- Hume's eLLM (empathic Large Language Model) is a multimodal language
1862
- model that takes into account both expression measures and language.
1863
- The eLLM generates short, empathic language responses and guides
1864
- text-to-speech (TTS) prosody.
1865
- tools:
1866
- type: optional<list<optional<ReturnUserDefinedTool>>>
1867
- docs: List of user-defined tools associated with this Config.
1868
1885
  builtin_tools:
1869
1886
  type: optional<list<optional<ReturnBuiltinTool>>>
1870
1887
  docs: List of built-in tools associated with this Config.
1871
1888
  event_messages: optional<ReturnEventMessageSpecs>
1872
- timeouts: optional<ReturnTimeoutSpecs>
1873
- nudges: optional<ReturnNudgeSpec>
1874
- webhooks:
1875
- type: optional<list<optional<ReturnWebhookSpec>>>
1876
- docs: Map of webhooks associated with this config.
1889
+ version_description:
1890
+ type: optional<string>
1891
+ docs: An optional description of the Config version.
1877
1892
  source:
1878
1893
  openapi: evi-openapi.json
1879
1894
  ReturnPagedChatsPaginationDirection:
@@ -2405,6 +2420,16 @@ types:
2405
2420
  type: list<ReturnChatAudioReconstruction>
2406
2421
  source:
2407
2422
  openapi: evi-openapi.json
2423
+ ReturnPromptVersionType:
2424
+ enum:
2425
+ - FIXED
2426
+ - LATEST
2427
+ docs: >-
2428
+ Versioning method for a Prompt. Either `FIXED` for using a fixed version
2429
+ number or `LATEST` for auto-updating to the latest version.
2430
+ inline: true
2431
+ source:
2432
+ openapi: evi-openapi.json
2408
2433
  PostedConfigPromptSpec:
2409
2434
  docs: >-
2410
2435
  Identifies which prompt to use in a a config OR how to create a new prompt
@@ -2571,10 +2596,12 @@ types:
2571
2596
  properties:
2572
2597
  enabled:
2573
2598
  type: optional<boolean>
2574
- docs: EVI will nudge user after inactivity
2599
+ docs: >-
2600
+ If true, EVI will 'nudge' the user to speak after a determined
2601
+ interval of silence.
2575
2602
  interval_secs:
2576
2603
  type: optional<integer>
2577
- docs: Time interval in seconds after which the nudge will be sent.
2604
+ docs: The interval of inactivity (in seconds) before a nudge is triggered.
2578
2605
  source:
2579
2606
  openapi: evi-openapi.json
2580
2607
  PostedTimeoutSpecsInactivity:
@@ -2693,6 +2720,15 @@ types:
2693
2720
  type: list<PostedWebhookEventType>
2694
2721
  source:
2695
2722
  openapi: evi-openapi.json
2723
+ ReturnVoiceSpec:
2724
+ discriminated: false
2725
+ docs: An voice specification posted to the server
2726
+ union:
2727
+ - ReturnVoiceSpecEvi2
2728
+ - ReturnVoiceSpecOctavecustom
2729
+ - ReturnVoiceSpecOctaveshared
2730
+ source:
2731
+ openapi: evi-openapi.json
2696
2732
  ReturnLanguageModel:
2697
2733
  docs: A specific LanguageModel
2698
2734
  properties:
@@ -3174,6 +3210,48 @@ types:
3174
3210
  docs: Duration in seconds for the timeout.
3175
3211
  source:
3176
3212
  openapi: evi-openapi.json
3213
+ ReturnVoiceSpecEvi2:
3214
+ docs: A specific voice specification for use with EVI 1 or 2
3215
+ properties:
3216
+ provider:
3217
+ type: string
3218
+ docs: The provider of the voice to use. Based on the enum VoiceProvider.
3219
+ name:
3220
+ type: optional<string>
3221
+ docs: >-
3222
+ String with the name of the voice to use. Maximum length of 75
3223
+ characters. Will be converted to all-uppercase.
3224
+ custom_voice: ReturnCustomVoice
3225
+ source:
3226
+ openapi: evi-openapi.json
3227
+ ReturnVoiceSpecOctavecustom:
3228
+ docs: A specific voice specification for an Octave voice
3229
+ properties:
3230
+ provider:
3231
+ type: string
3232
+ docs: The provider of the voice to use. Based on the enum VoiceProvider.
3233
+ name:
3234
+ type: optional<string>
3235
+ docs: >-
3236
+ String with the name of the voice to use. Maximum length of 75
3237
+ characters. Will be converted to all-uppercase.
3238
+ custom_voice: optional<ReturnVoiceOctaveCustom>
3239
+ source:
3240
+ openapi: evi-openapi.json
3241
+ ReturnVoiceSpecOctaveshared:
3242
+ docs: A specific voice specification for an Octave voice
3243
+ properties:
3244
+ provider:
3245
+ type: string
3246
+ docs: The provider of the voice to use. Based on the enum VoiceProvider.
3247
+ name:
3248
+ type: optional<string>
3249
+ docs: >-
3250
+ String with the name of the voice to use. Maximum length of 75
3251
+ characters. Will be converted to all-uppercase.
3252
+ custom_voice: optional<ReturnVoiceOctaveShared>
3253
+ source:
3254
+ openapi: evi-openapi.json
3177
3255
  ReturnEventMessageSpec:
3178
3256
  docs: A specific event message configuration to be returned from the server
3179
3257
  properties:
@@ -3213,6 +3291,53 @@ types:
3213
3291
  minutes).
3214
3292
  source:
3215
3293
  openapi: evi-openapi.json
3294
+ ReturnVoiceOctaveCustom:
3295
+ docs: An octave custom voice specification returned from the server
3296
+ properties:
3297
+ id: optional<string>
3298
+ version: optional<integer>
3299
+ name: optional<string>
3300
+ description: optional<string>
3301
+ voice_type: optional<string>
3302
+ created_on: optional<long>
3303
+ modified_on: optional<long>
3304
+ voice_prompt: optional<string>
3305
+ normalization_gain_db: optional<double>
3306
+ speaker_embedding: optional<string>
3307
+ reference_uri: optional<string>
3308
+ reference_audio_format: optional<string>
3309
+ clone_reference_uri: optional<string>
3310
+ clone_reference_audio_format: optional<string>
3311
+ reference_tokens: optional<string>
3312
+ reference_atoks: optional<string>
3313
+ reference_decoder_dtoks: optional<string>
3314
+ reference_text: optional<string>
3315
+ signed_preview_reference_uri: optional<ReturnSignedUri>
3316
+ signed_clone_reference_uri: optional<ReturnSignedUri>
3317
+ image_uri: optional<string>
3318
+ disabled: optional<boolean>
3319
+ source:
3320
+ openapi: evi-openapi.json
3321
+ ReturnVoiceOctaveShared:
3322
+ docs: An octave shared voice specification returned from the server
3323
+ properties:
3324
+ id: optional<string>
3325
+ version: optional<integer>
3326
+ name: optional<string>
3327
+ description: optional<string>
3328
+ reference_signed_uri: optional<ReturnSignedUri>
3329
+ image_uri: optional<string>
3330
+ source:
3331
+ openapi: evi-openapi.json
3332
+ ReturnSignedUri:
3333
+ docs: A signed URI linking to a file in GCS
3334
+ properties:
3335
+ filename: optional<string>
3336
+ method: optional<string>
3337
+ signed_uri: optional<string>
3338
+ expiration_timestamp_millis: optional<long>
3339
+ source:
3340
+ openapi: evi-openapi.json
3216
3341
  PostedVoiceProvider:
3217
3342
  enum:
3218
3343
  - HUME_AI
@@ -47,14 +47,6 @@ channel:
47
47
 
48
48
  Include this parameter to apply a specific version of an EVI
49
49
  configuration. If omitted, the latest version will be applied.
50
- enable_context_injection:
51
- type: optional<boolean>
52
- default: true
53
- docs: >-
54
- A flag to enable context injection. Set this query parameter to true to
55
- have EVI inject context into the conversation. This is useful for
56
- scenarios where you want to provide additional information to EVI to
57
- help it better understand the user's request.
58
50
  event_limit:
59
51
  type: optional<integer>
60
52
  docs: >-
@@ -164,6 +156,7 @@ types:
164
156
  union:
165
157
  - type: root.AssistantEnd
166
158
  - type: root.AssistantMessage
159
+ - type: root.AssistantProsody
167
160
  - type: root.AudioOutput
168
161
  - type: root.ChatMetadata
169
162
  - type: root.WebSocketError
@@ -202,7 +202,13 @@ service:
202
202
  type: optional<list<optional<root.PostedBuiltinTool>>>
203
203
  docs: List of built-in tools associated with this Config.
204
204
  event_messages: optional<root.PostedEventMessageSpecs>
205
- nudges: optional<root.PostedNudgeSpec>
205
+ nudges:
206
+ type: optional<root.PostedNudgeSpec>
207
+ docs: >-
208
+ Configures nudges, brief audio prompts that can guide
209
+ conversations when users pause or need encouragement to continue
210
+ speaking. Nudges help create more natural, flowing interactions
211
+ by providing gentle conversational cues.
206
212
  timeouts: optional<root.PostedTimeoutSpecs>
207
213
  webhooks:
208
214
  type: optional<list<optional<root.PostedWebhookSpec>>>
@@ -392,22 +392,19 @@ types:
392
392
  see our documentation on [instant
393
393
  mode](/docs/text-to-speech-tts/overview#ultra-low-latency-streaming-instant-mode).
394
394
 
395
- - Dynamic voice generation is not supported with this mode; a
396
- predefined
395
+ - A
397
396
  [voice](/reference/text-to-speech-tts/synthesize-json-streaming#request.body.utterances.voice)
398
- must be specified in your request.
397
+ must be specified when instant mode is enabled. Dynamic voice
398
+ generation is not supported with this mode.
399
399
 
400
- - This mode is only supported for streaming endpoints (e.g.,
400
+ - Instant mode is only supported for streaming endpoints (e.g.,
401
401
  [/v0/tts/stream/json](/reference/text-to-speech-tts/synthesize-json-streaming),
402
402
  [/v0/tts/stream/file](/reference/text-to-speech-tts/synthesize-file-streaming)).
403
403
 
404
404
  - Ensure only a single generation is requested
405
405
  ([num_generations](/reference/text-to-speech-tts/synthesize-json-streaming#request.body.num_generations)
406
406
  must be `1` or omitted).
407
-
408
- - With `instant_mode` enabled, **requests incur a 10% higher cost**
409
- due to increased compute and resource requirements.
410
- default: false
407
+ default: true
411
408
  source:
412
409
  openapi: tts-openapi.yml
413
410
  ReturnTts:
@@ -515,14 +512,20 @@ types:
515
512
  docs: >-
516
513
  Natural language instructions describing how the synthesized speech
517
514
  should sound, including but not limited to tone, intonation, pacing,
518
- and accent (e.g., 'a soft, gentle voice with a strong British
519
- accent').
515
+ and accent.
520
516
 
521
- - If a Voice is specified in the request, this description serves as
522
- acting instructions. For tips on how to effectively guide speech
523
- delivery, see our guide on [Acting
517
+
518
+ **This field behaves differently depending on whether a voice is
519
+ specified**:
520
+
521
+ - **Voice specified**: the description will serve as acting directions
522
+ for delivery. Keep directions concise—100 characters or fewer—for best
523
+ results. See our guide on [acting
524
524
  instructions](/docs/text-to-speech-tts/acting-instructions).
525
- - If no Voice is specified, a new voice is generated based on this description. See our [prompting guide](/docs/text-to-speech-tts/prompting) for tips on designing a voice.
525
+
526
+ - **Voice not specified**: the description will serve as a voice
527
+ prompt for generating a voice. See our [prompting
528
+ guide](/docs/text-to-speech-tts/prompting) for design tips.
526
529
  validation:
527
530
  maxLength: 1000
528
531
  speed:
package/Client.js CHANGED
@@ -45,8 +45,8 @@ const fetcherThatAddsHeaders = (fetcherToWrap) => {
45
45
  var _a;
46
46
  const newArgs = Object.assign({}, args);
47
47
  newArgs.headers = (_a = newArgs.headers) !== null && _a !== void 0 ? _a : {};
48
- (newArgs.headers["X-Hume-Client-Name"] = "typescript_sdk"),
49
- (newArgs.headers["X-Hume-Client-Version"] = version_1.SDK_VERSION);
48
+ ((newArgs.headers["X-Hume-Client-Name"] = "typescript_sdk"),
49
+ (newArgs.headers["X-Hume-Client-Version"] = version_1.SDK_VERSION));
50
50
  return fetcherToWrap(args);
51
51
  };
52
52
  };
@@ -2,4 +2,4 @@
2
2
  * This file was auto-generated by Fern from our API Definition.
3
3
  */
4
4
  import * as Hume from "../../../../../index";
5
- export type SubscribeEvent = Hume.empathicVoice.AssistantEnd | Hume.empathicVoice.AssistantMessage | Hume.empathicVoice.AudioOutput | Hume.empathicVoice.ChatMetadata | Hume.empathicVoice.WebSocketError | Hume.empathicVoice.UserInterruption | Hume.empathicVoice.UserMessage | Hume.empathicVoice.ToolCallMessage | Hume.empathicVoice.ToolResponseMessage | Hume.empathicVoice.ToolErrorMessage;
5
+ export type SubscribeEvent = Hume.empathicVoice.AssistantEnd | Hume.empathicVoice.AssistantMessage | Hume.empathicVoice.AssistantProsody | Hume.empathicVoice.AudioOutput | Hume.empathicVoice.ChatMetadata | Hume.empathicVoice.WebSocketError | Hume.empathicVoice.UserInterruption | Hume.empathicVoice.UserMessage | Hume.empathicVoice.ToolCallMessage | Hume.empathicVoice.ToolResponseMessage | Hume.empathicVoice.ToolErrorMessage;
@@ -63,6 +63,7 @@ export interface PostedConfig {
63
63
  /** List of built-in tools associated with this Config. */
64
64
  builtinTools?: (Hume.empathicVoice.PostedBuiltinTool | undefined)[];
65
65
  eventMessages?: Hume.empathicVoice.PostedEventMessageSpecs;
66
+ /** Configures nudges, brief audio prompts that can guide conversations when users pause or need encouragement to continue speaking. Nudges help create more natural, flowing interactions by providing gentle conversational cues. */
66
67
  nudges?: Hume.empathicVoice.PostedNudgeSpec;
67
68
  timeouts?: Hume.empathicVoice.PostedTimeoutSpecs;
68
69
  /** Webhook config specifications for each subscriber. */
@@ -0,0 +1,21 @@
1
+ /**
2
+ * This file was auto-generated by Fern from our API Definition.
3
+ */
4
+ import * as Hume from "../../../index";
5
+ /**
6
+ * When provided, the output is an Assistant Prosody message.
7
+ */
8
+ export interface AssistantProsody {
9
+ /** Used to manage conversational state, correlate frontend and backend data, and persist conversations across EVI sessions. */
10
+ customSessionId?: string;
11
+ /** Unique identifier for the segment. */
12
+ id?: string;
13
+ /** Inference model results. */
14
+ models: Hume.empathicVoice.Inference;
15
+ /**
16
+ * The type of message sent through the socket; for an Assistant Prosody message, this must be `assistant_PROSODY`.
17
+ *
18
+ * This message the expression measurement predictions of the assistant's audio output.
19
+ */
20
+ type: "assistant_prosody";
21
+ }
@@ -0,0 +1,5 @@
1
+ "use strict";
2
+ /**
3
+ * This file was auto-generated by Fern from our API Definition.
4
+ */
5
+ Object.defineProperty(exports, "__esModule", { value: true });
@@ -5,8 +5,8 @@
5
5
  * A nudge specification posted to the server
6
6
  */
7
7
  export interface PostedNudgeSpec {
8
- /** EVI will nudge user after inactivity */
8
+ /** If true, EVI will 'nudge' the user to speak after a determined interval of silence. */
9
9
  enabled?: boolean;
10
- /** Time interval in seconds after which the nudge will be sent. */
10
+ /** The interval of inactivity (in seconds) before a nudge is triggered. */
11
11
  intervalSecs?: number;
12
12
  }
@@ -6,6 +6,8 @@ import * as Hume from "../../../index";
6
6
  * A specific config version returned from the server
7
7
  */
8
8
  export interface ReturnConfig {
9
+ /** Name applied to all versions of a particular Config. */
10
+ name?: string;
9
11
  /** Identifier for a Config. Formatted as a UUID. */
10
12
  id?: string;
11
13
  /**
@@ -18,36 +20,34 @@ export interface ReturnConfig {
18
20
  version?: number;
19
21
  /** Specifies the EVI version to use. Use `"1"` for version 1, or `"2"` for the latest enhanced version. For a detailed comparison of the two versions, refer to our [guide](/docs/empathic-voice-interface-evi/configuration/evi-version). */
20
22
  eviVersion?: string;
21
- /** An optional description of the Config version. */
22
- versionDescription?: string;
23
- /** Name applied to all versions of a particular Config. */
24
- name?: string;
23
+ timeouts?: Hume.empathicVoice.ReturnTimeoutSpecs;
24
+ nudges?: Hume.empathicVoice.ReturnNudgeSpec;
25
+ /**
26
+ * The eLLM setup associated with this Config.
27
+ *
28
+ * Hume's eLLM (empathic Large Language Model) is a multimodal language model that takes into account both expression measures and language. The eLLM generates short, empathic language responses and guides text-to-speech (TTS) prosody.
29
+ */
30
+ ellmModel?: Hume.empathicVoice.ReturnEllmModel;
31
+ /** A voice specification associated with this Config. */
32
+ voice?: Hume.empathicVoice.ReturnVoice;
33
+ prompt?: Hume.empathicVoice.ReturnPrompt;
34
+ /** List of user-defined tools associated with this Config. */
35
+ tools?: (Hume.empathicVoice.ReturnUserDefinedTool | undefined)[];
36
+ /** Map of webhooks associated with this config. */
37
+ webhooks?: (Hume.empathicVoice.ReturnWebhookSpec | undefined)[];
25
38
  /** Time at which the Config was created. Measured in seconds since the Unix epoch. */
26
39
  createdOn?: number;
27
40
  /** Time at which the Config was last modified. Measured in seconds since the Unix epoch. */
28
41
  modifiedOn?: number;
29
- prompt?: Hume.empathicVoice.ReturnPrompt;
30
- /** A voice specification associated with this Config. */
31
- voice?: Hume.empathicVoice.ReturnVoice;
32
42
  /**
33
43
  * The supplemental language model associated with this Config.
34
44
  *
35
45
  * This model is used to generate longer, more detailed responses from EVI. Choosing an appropriate supplemental language model for your use case is crucial for generating fast, high-quality responses from EVI.
36
46
  */
37
47
  languageModel?: Hume.empathicVoice.ReturnLanguageModel;
38
- /**
39
- * The eLLM setup associated with this Config.
40
- *
41
- * Hume's eLLM (empathic Large Language Model) is a multimodal language model that takes into account both expression measures and language. The eLLM generates short, empathic language responses and guides text-to-speech (TTS) prosody.
42
- */
43
- ellmModel?: Hume.empathicVoice.ReturnEllmModel;
44
- /** List of user-defined tools associated with this Config. */
45
- tools?: (Hume.empathicVoice.ReturnUserDefinedTool | undefined)[];
46
48
  /** List of built-in tools associated with this Config. */
47
49
  builtinTools?: (Hume.empathicVoice.ReturnBuiltinTool | undefined)[];
48
50
  eventMessages?: Hume.empathicVoice.ReturnEventMessageSpecs;
49
- timeouts?: Hume.empathicVoice.ReturnTimeoutSpecs;
50
- nudges?: Hume.empathicVoice.ReturnNudgeSpec;
51
- /** Map of webhooks associated with this config. */
52
- webhooks?: (Hume.empathicVoice.ReturnWebhookSpec | undefined)[];
51
+ /** An optional description of the Config version. */
52
+ versionDescription?: string;
53
53
  }
@@ -6,8 +6,18 @@ import * as Hume from "../../../index";
6
6
  * A Prompt associated with this Config.
7
7
  */
8
8
  export interface ReturnPrompt {
9
+ /** Name applied to all versions of a particular Prompt. */
10
+ name: string;
9
11
  /** Identifier for a Prompt. Formatted as a UUID. */
10
12
  id: string;
13
+ /**
14
+ * Instructions used to shape EVI’s behavior, responses, and style.
15
+ *
16
+ * You can use the Prompt to define a specific goal or role for EVI, specifying how it should act or what it should focus on during the conversation. For example, EVI can be instructed to act as a customer support representative, a fitness coach, or a travel advisor, each with its own set of behaviors and response styles.
17
+ *
18
+ * For help writing a system prompt, see our [Prompting Guide](/docs/empathic-voice-interface-evi/guides/prompting).
19
+ */
20
+ text: string;
11
21
  /**
12
22
  * Version number for a Prompt.
13
23
  *
@@ -18,20 +28,10 @@ export interface ReturnPrompt {
18
28
  version: number;
19
29
  /** Versioning method for a Prompt. Either `FIXED` for using a fixed version number or `LATEST` for auto-updating to the latest version. */
20
30
  versionType: Hume.empathicVoice.ReturnPromptVersionType;
21
- /** An optional description of the Prompt version. */
22
- versionDescription?: string;
23
- /** Name applied to all versions of a particular Prompt. */
24
- name: string;
25
31
  /** Time at which the Prompt was created. Measured in seconds since the Unix epoch. */
26
32
  createdOn: number;
27
33
  /** Time at which the Prompt was last modified. Measured in seconds since the Unix epoch. */
28
34
  modifiedOn: number;
29
- /**
30
- * Instructions used to shape EVI’s behavior, responses, and style.
31
- *
32
- * You can use the Prompt to define a specific goal or role for EVI, specifying how it should act or what it should focus on during the conversation. For example, EVI can be instructed to act as a customer support representative, a fitness coach, or a travel advisor, each with its own set of behaviors and response styles.
33
- *
34
- * For help writing a system prompt, see our [Prompting Guide](/docs/empathic-voice-interface-evi/guides/prompting).
35
- */
36
- text: string;
35
+ /** An optional description of the Prompt version. */
36
+ versionDescription?: string;
37
37
  }
@@ -0,0 +1,12 @@
1
+ /**
2
+ * This file was auto-generated by Fern from our API Definition.
3
+ */
4
+ /**
5
+ * A signed URI linking to a file in GCS
6
+ */
7
+ export interface ReturnSignedUri {
8
+ filename?: string;
9
+ method?: string;
10
+ signedUri?: string;
11
+ expirationTimestampMillis?: number;
12
+ }
@@ -0,0 +1,5 @@
1
+ "use strict";
2
+ /**
3
+ * This file was auto-generated by Fern from our API Definition.
4
+ */
5
+ Object.defineProperty(exports, "__esModule", { value: true });
@@ -0,0 +1,31 @@
1
+ /**
2
+ * This file was auto-generated by Fern from our API Definition.
3
+ */
4
+ import * as Hume from "../../../index";
5
+ /**
6
+ * An octave custom voice specification returned from the server
7
+ */
8
+ export interface ReturnVoiceOctaveCustom {
9
+ id?: string;
10
+ version?: number;
11
+ name?: string;
12
+ description?: string;
13
+ voiceType?: string;
14
+ createdOn?: number;
15
+ modifiedOn?: number;
16
+ voicePrompt?: string;
17
+ normalizationGainDb?: number;
18
+ speakerEmbedding?: string;
19
+ referenceUri?: string;
20
+ referenceAudioFormat?: string;
21
+ cloneReferenceUri?: string;
22
+ cloneReferenceAudioFormat?: string;
23
+ referenceTokens?: string;
24
+ referenceAtoks?: string;
25
+ referenceDecoderDtoks?: string;
26
+ referenceText?: string;
27
+ signedPreviewReferenceUri?: Hume.empathicVoice.ReturnSignedUri;
28
+ signedCloneReferenceUri?: Hume.empathicVoice.ReturnSignedUri;
29
+ imageUri?: string;
30
+ disabled?: boolean;
31
+ }