|
9 | 9 | "\n",
|
10 | 10 | "<img\n",
|
11 | 11 | " src=\"../images/realtime_prompting_guide.png\"\n",
|
12 |
| - " style=\"width:450px; height:auto;\"\n", |
| 12 | + " style=\"width:450px; height:450px;\"\n", |
13 | 13 | "/>\n",
|
14 | 14 | "\n",
|
15 | 15 | "\n",
|
|
20 | 20 | "\n",
|
21 | 21 | "The new gpt-realtime model delivers stronger instruction following, more reliable tool calling, noticeably better voice quality, and an overall smoother feel. These gains make it practical to move from chained approaches to true realtime experiences, cutting latency and producing responses that sound more natural and expressive.\n",
|
22 | 22 | "\n",
|
23 |
| - "Realtime model benefits from different prompting techniques that wouldn't directly apply to text based models. This prompting guide starts with a suggested prompt skeleton, then walks through each part with practical tips, small patterns you can copy, and examples you can adapt to your use case.\n", |
24 |
| - "\n", |
25 |
| - "# Table of Contents\n", |
26 |
| - "\n", |
27 |
| - "- [Realtime Prompting Guide](#realtime-prompting-guide)\n", |
28 |
| - "- [General Tips](#general-tips)\n", |
29 |
| - "- [Prompt Structure](#prompt-structure)\n", |
30 |
| - "- [Role and Objective](#role-and-objective)\n", |
31 |
| - "- [Personality and Tone](#personality-and-tone)\n", |
32 |
| - " - [Speed Instructions](#speed-instructions)\n", |
33 |
| - " - [Language Constraint](#language-constraint)\n", |
34 |
| - " - [Reduce Repetition](#reduce-repetition)\n", |
35 |
| - "- [Reference Pronunciations](#reference-pronunciations)\n", |
36 |
| - " - [Alphanumeric Pronunciations](#alphanumeric-pronunciations)\n", |
37 |
| - "- [Instructions](#instructions)\n", |
38 |
| - " - [Instruction Following](#instruction-following)\n", |
39 |
| - " - [No audio or unclear audio](#no-audio-or-unclear-audio)\n", |
40 |
| - "- [Tools](#tools)\n", |
41 |
| - " - [Tool Selection](#tool-selection)\n", |
42 |
| - " - [Tool Call Preambles](#tool-call-preambles)\n", |
43 |
| - " - [Tool Call Preambles + Sample Phrases](#tool-call-preambles-sample-phrases)\n", |
44 |
| - " - [Tool Calls without Confirmation](#tool-calls-without-confirmation)\n", |
45 |
| - " - [Tool Call Performance](#tool-call-performance)\n", |
46 |
| - " - [Tool Level Behavior](#tool-level-behavior)\n", |
47 |
| - " - [Rephrase Supervisor Tool (Responder-Thinker Architecture)](#rephrase-supervisor-tool-responder-thinker-architecture)\n", |
48 |
| - " - [Common Tools](#common-tools)\n", |
49 |
| - "- [Conversation flow](#conversation-flow)\n", |
50 |
| - " - [Sample phrases](#sample-phrases)\n", |
51 |
| - " - [Conversation flow + Sample Phrases](#conversation-flow-sample-phrases)\n", |
52 |
| - " - [Advanced Conversation Flow](#advanced-conversation-flow)\n", |
53 |
| - " - [Conversation Flow as State Machine](#conversation-flow-as-state-machine)\n", |
54 |
| - " - [Dynamic Conversation Flow](#dynamic-conversation-flow)\n", |
55 |
| - "- [Safety & Escalation](#safety-escalation)" |
| 23 | + "Realtime model benefits from different prompting techniques that wouldn't directly apply to text based models. This prompting guide starts with a suggested prompt skeleton, then walks through each part with practical tips, small patterns you can copy, and examples you can adapt to your use case." |
56 | 24 | ]
|
57 | 25 | },
|
58 | 26 | {
|
|
0 commit comments