Skip to main content
deepgram-sdk-patterns Apply production-ready Deepgram SDK patterns for TypeScript and Python.
Use when implementing Deepgram integrations, refactoring SDK usage,
or establishing team coding standards for Deepgram.
Trigger: "deepgram SDK patterns", "deepgram best practices",
"deepgram code patterns", "idiomatic deepgram", "deepgram typescript".
跳到安装 Skills Marketplace 发现并探索由社区构建的 Agent Skills
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/jeremylongshore/claude-code-plugins-plus-skills --skill deepgram-sdk-patterns命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
下载 Zip 下载中... jeremylongshore
jeremylongshore/claude-code-plugins-plus-skills
打开 GitHub 仓库 同仓库更多 Skills Implement user sign-up and sign-in flows with Clerk.
Use when building authentication UI, customizing sign-in experience,
or implementing OAuth social login.
Trigger with phrases like "clerk sign-in", "clerk sign-up",
"clerk login flow", "clerk OAuth", "clerk social login".
Implement session management and middleware with Clerk.
Use when managing user sessions, configuring route protection,
or implementing token refresh and custom JWT templates.
Trigger with phrases like "clerk session", "clerk middleware",
"clerk route protection", "clerk token", "clerk JWT".
Configure enterprise SSO, role-based access control, and organization management.
Use when implementing SSO integration, configuring role-based permissions,
or setting up organization-level controls.
Trigger with phrases like "clerk SSO", "clerk RBAC",
"clerk enterprise", "clerk roles", "clerk permissions", "clerk organizations".
name deepgram-sdk-patterns description Apply production-ready Deepgram SDK patterns for TypeScript and Python.
Use when implementing Deepgram integrations, refactoring SDK usage,
or establishing team coding standards for Deepgram.
Trigger: "deepgram SDK patterns", "deepgram best practices",
"deepgram code patterns", "idiomatic deepgram", "deepgram typescript".
allowed-tools Read, Write, Edit version 1.13.0 license MIT author Jeremy Longshore <jeremy@intentsolutions.io> tags ["saas","deepgram","python","typescript","patterns"] compatibility Designed for Claude Code, also compatible with Codex and OpenClaw
Deepgram SDK Patterns
Overview
Production patterns for @deepgram/sdk (TypeScript) and deepgram-sdk (Python). Covers singleton client, typed wrappers, text-to-speech with Aura, audio intelligence pipeline, error handling, and SDK v5 migration path.
Prerequisites
npm install @deepgram/sdk or pip install deepgram-sdkDEEPGRAM_API_KEY environment variable configured
Instructions
Step 1: Singleton Client (TypeScript)
import { createClient, DeepgramClient } from '@deepgram/sdk' ;
class DeepgramService {
private static instance : DeepgramService ;
private client : DeepgramClient ;
private constructor (const apiKey = process.env .DEEPGRAM_API_KEY ;
if (!apiKey) throw new Error ('DEEPGRAM_API_KEY is required' );
this .client = createClient (apiKey);
}
static getInstance (): DeepgramService {
if (!this .instance ) this .instance = new DeepgramService ();
return this .instance
getClient
DeepgramClient
return
this
client
export
const
DeepgramService
getInstance
getClient
Step 2: Text-to-Speech with Aura import { createClient } from '@deepgram/sdk' ;
import { writeFileSync } from 'fs' ;
const deepgram = createClient (process.env .DEEPGRAM_API_KEY !);
async function textToSpeech (text : string , outputPath : string const response = await deepgram.speak .request (
{ text },
{
model : 'aura-2-thalia-en' ,
encoding : 'linear16' ,
container : 'wav' ,
sample_rate : 24000 ,
}
);
const stream = await response.getStream ();
if (!stream) throw new Error ('No audio stream returned' );
const reader = stream.getReader ();
const chunks : Uint8Array [] = [];
while (true ) {
const { done, value } = await reader.read ();
if (done) break ;
chunks.push (value);
}
const buffer = Buffer .concat (chunks);
writeFileSync (outputPath, buffer);
console .log (`Audio saved: ${outputPath} (${buffer.length} bytes)` );
return buffer;
}
Step 3: Audio Intelligence Pipeline async function analyzeConversation (audioUrl : string const { result, error } = await deepgram.listen .prerecorded .transcribeUrl (
{ url : audioUrl },
{
model : 'nova-3' ,
smart_format : true ,
diarize : true ,
utterances : true ,
summarize : 'v2' ,
detect_topics : true ,
sentiment : true ,
intents : true ,
}
);
if (error) throw error;
return {
transcript : result.results .channels [0 ].alternatives [0 ].transcript ,
summary : result.results .summary ?.short ,
topics : result.results .topics ?.segments ?.map ((s : any ({
text : s.text ,
topics : s.topics .map ((t : any t.topic ),
})),
sentiments : result.results .sentiments ?.segments ?.map ((s : any ({
text : s.text ,
sentiment : s.sentiment ,
confidence : s.sentiment_score ,
})),
intents : result.results .intents ?.segments ?.map ((s : any ({
text : s.text ,
intent : s.intents [0 ]?.intent ,
confidence : s.intents [0 ]?.confidence_score ,
})),
};
}
Step 4: Python Production Patterns from deepgram import DeepgramClient, PrerecordedOptions, LiveOptions, SpeakOptions
import os
class DeepgramService :
_instance = None
def __new__ (cls ):
if cls._instance is None :
cls._instance = super ().__new__(cls)
cls._instance.client = DeepgramClient(os.environ["DEEPGRAM_API_KEY" ])
return cls._instance
def transcribe_url (self, url: str , **kwargs ):
options = PrerecordedOptions(
model=kwargs.get("model" , "nova-3" ),
smart_format=True ,
diarize=kwargs.get("diarize" , False ),
summarize=kwargs.get("summarize" , False ),
)
source = {"url" : url}
return self .client.listen.rest.v("1" ).transcribe_url(source, options)
def transcribe_file (self, path: str , **kwargs ):
with open (path, "rb" ) as f:
source = {"buffer" : f.read(), "mimetype" : self ._mimetype(path)}
options = PrerecordedOptions(
model=kwargs.get("model" , "nova-3" ),
smart_format=True ,
diarize=kwargs.get("diarize" , False ),
)
return self .client.listen.rest.v("1" ).transcribe_file(source, options)
def text_to_speech (self, text: str , output_path: str ):
options = SpeakOptions(model="aura-2-thalia-en" , encoding="linear16" )
response = self .client.speak.rest.v("1" ).save(output_path, {"text" : text}, options)
return response
@staticmethod
def _mimetype (path: str ) -> str :
ext = path.rsplit("." , 1 )[-1 ].lower()
return {"wav" : "audio/wav" , "mp3" : "audio/mpeg" , "flac" : "audio/flac" ,
"ogg" : "audio/ogg" , "m4a" : "audio/mp4" }.get(ext, "audio/wav" )
Step 5: Typed Response Helpers
interface TranscriptWord {
word : string ;
start : number ;
end : number ;
confidence : number ;
speaker ?: number ;
punctuated_word ?: string ;
}
interface TranscriptResult {
transcript : string ;
confidence : number ;
words : TranscriptWord [];
duration : number ;
requestId : string ;
}
function parseResult (result : any TranscriptResult {
const alt = result.results .channels [0 ].alternatives [0 ];
return {
transcript : alt.transcript ,
confidence : alt.confidence ,
words : alt.words ?? [],
duration : result.metadata .duration ,
requestId : result.metadata .request_id ,
};
}
Step 6: SDK v5 Migration Notes
import { createClient } from '@deepgram/sdk' ;
const dg = createClient (apiKey);
await dg.listen .prerecorded .transcribeUrl (source, options);
await dg.listen .live (options);
await dg.speak .request ({ text }, options);
import { DeepgramClient } from '@deepgram/sdk' ;
const dg = new DeepgramClient ({ apiKey });
await dg.listen .v1 .media .transcribeUrl (source, options);
await dg.listen .v1 .connect (options);
await dg.speak .v1 .audio .generate ({ text }, options);
Output
Singleton client pattern with environment validation
Text-to-speech (Aura-2) with stream-to-file
Audio intelligence pipeline (summary, topics, sentiment, intents)
Python production service class
Typed response helpers
v5 migration reference
Error Handling Error Cause Solution 401 UnauthorizedInvalid API key Check DEEPGRAM_API_KEY value 400 Unsupported formatBad audio codec Convert to WAV/MP3/FLAC speak.request is not a functionSDK version mismatch Check import, v5 uses speak.v1.audio.generate Empty TTS response Empty text input Validate text is non-empty before calling summarize returns nullFeature not enabled Pass summarize: 'v2' (string, not boolean)
Resources
Next Steps Proceed to deepgram-data-handling for transcript storage and processing patterns.
