Skip to content

Commit e36ec6d

Browse files
kristinaorekhovagitbook-bot
kristinaorekhova
authored and
gitbook-bot
committed
GITBOOK-1061: No subject
1 parent be604d4 commit e36ec6d

4 files changed

Lines changed: 566 additions & 0 deletions

File tree

‎docs/SUMMARY.md‎

Lines changed: 3 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -432,6 +432,9 @@
432432
* [slam-1](api-references/speech-models/speech-to-text/assembly-ai/slam-1.md)
433433
* [universal](api-references/speech-models/speech-to-text/assembly-ai/universal.md)
434434
* [Deepgram](api-references/speech-voice-models/stt/Deepgram/README.md)
435+
* [Nova 3](api-references/speech-models/speech-to-text/deepgram/nova-3.md)
436+
* [Nova 3 Medical](api-references/speech-models/speech-to-text/deepgram/nova-3-medical.md)
437+
* [Nova 3 General](api-references/speech-models/speech-to-text/deepgram/nova-3-general.md)
435438
* [nova-2](api-references/speech-voice-models/stt/Deepgram/nova-2.md)
436439
* [OpenAI](api-references/speech-voice-models/stt/OpenAI/README.md)
437440
* [whisper-base](api-references/speech-voice-models/stt/OpenAI/whisper-base.md)

‎docs/api-references/speech-models/speech-to-text/deepgram/nova-3-general.md‎

Lines changed: 189 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,189 @@
1+
---
2+
layout:
3+
width: wide
4+
title:
5+
visible: true
6+
description:
7+
visible: true
8+
tableOfContents:
9+
visible: true
10+
outline:
11+
visible: true
12+
pagination:
13+
visible: true
14+
metadata:
15+
visible: true
16+
tags:
17+
visible: true
18+
actions:
19+
visible: true
20+
---
21+
22+
# Nova 3 General
23+
24+
{% columns %}
25+
{% column width="66.66666666666666%" %}
26+
{% hint style="info" %}
27+
This documentation is valid for the following list of our models:
28+
29+
* `nova-3-general`
30+
{% endhint %}
31+
{% endcolumn %}
32+
33+
{% column width="33.33333333333334%" %}
34+
<a href="https://aimlapi.com/app/deepgram/nova-3-general" class="button primary">Try in Playground</a>
35+
{% endcolumn %}
36+
{% endcolumns %}
37+
38+
Nova-3 General — multilingual speech-to-text model from Deepgram with automatic language detection across 30+ languages, optimized for real-time and batch transcription with speaker diarization and advanced audio intelligence features.
39+
40+
{% hint style="success" %}
41+
This model uses per-minute billing. The cost of audio transcription is based on the duration of the input audio file.
42+
{% endhint %}
43+
44+
## Setup your API Key
45+
46+
If you don't have an API key for the AI/ML API yet, feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up).
47+
48+
## API Schemas
49+
50+
{% openapi-operation spec="deepgram-nova-3-general" path="/v1/stt/create" method="post" %}
51+
[OpenAPI deepgram-nova-3-general](https://raw.githubusercontent.com/aimlapi/api-docs/refs/heads/main/docs/api-references/speech-models/Deepgram/nova-3-general.json)
52+
{% endopenapi-operation %}
53+
54+
## Quick Example: Processing a Speech Audio File via URL
55+
56+
Let's transcribe the following audio fragment:
57+
58+
{% embed url="https://drive.google.com/file/d/1ZN-28NUbK1TXHt6oEPj42zUJCv82e9L4/view?usp=sharing" %}
59+
60+
{% code overflow="wrap" %}
61+
```python
62+
import time
63+
import requests
64+
import json
65+
66+
base_url = "https://api.aimlapi.com/v1"
67+
# Insert your AIML API Key instead of <YOUR_AIMLAPI_KEY>:
68+
api_key = "<YOUR_AIMLAPI_KEY>"
69+
70+
# Creating and sending a speech-to-text conversion task to the server
71+
def create_stt():
72+
url = f"{base_url}/stt/create"
73+
headers = {
74+
"Authorization": f"Bearer {api_key}",
75+
}
76+
77+
data = {
78+
"model": "nova-3-general",
79+
"url": "https://audio-samples.github.io/samples/mp3/blizzard_primed/sample-0.mp3"
80+
}
81+
82+
response = requests.post(url, json=data, headers=headers)
83+
84+
if response.status_code >= 400:
85+
print(f"Error: {response.status_code} - {response.text}")
86+
else:
87+
response_data = response.json()
88+
print(response_data)
89+
return response_data
90+
91+
# Requesting the result of the task from the server using the generation_id
92+
def get_stt(gen_id):
93+
url = f"{base_url}/stt/{gen_id}"
94+
headers = {
95+
"Authorization": f"Bearer {api_key}",
96+
}
97+
response = requests.get(url, headers=headers)
98+
return response.json()
99+
100+
# First, start the generation, then repeatedly request the result from the server every 10 seconds.
101+
def main():
102+
stt_response = create_stt()
103+
gen_id = stt_response.get("generation_id")
104+
105+
if gen_id:
106+
start_time = time.time()
107+
timeout = 600
108+
while time.time() - start_time < timeout:
109+
response_data = get_stt(gen_id)
110+
111+
if response_data is None:
112+
print("Error: No response from API")
113+
break
114+
115+
status = response_data.get("status")
116+
117+
if status == "waiting" or status == "active":
118+
print("Still waiting... Checking again in 10 seconds.")
119+
time.sleep(10)
120+
else:
121+
print("Processing complete:\n", response_data["result"]["results"]["channels"][0]["alternatives"][0]["transcript"])
122+
123+
# Uncomment the line below to print the entire result object with all service data
124+
# print("Processing complete:\n", json.dumps(response_data["result"], indent=2, ensure_ascii=False))
125+
return response_data
126+
127+
print("Timeout reached. Stopping.")
128+
return None
129+
130+
131+
if __name__ == "__main__":
132+
main()
133+
```
134+
{% endcode %}
135+
136+
<details>
137+
138+
<summary>Response</summary>
139+
140+
{% code overflow="wrap" %}
141+
```json5
142+
{'generation_id': 'b2c3d4e5-f6a7-8901-bcde-f12345678901'}
143+
Still waiting... Checking again in 10 seconds.
144+
Processing complete:
145+
{
146+
"status": "completed",
147+
"result": {
148+
"metadata": {
149+
"request_id": "b2c3d4e5-f6a7-8901-bcde-f12345678901",
150+
"created": "2025-06-04T12:00:00.000Z",
151+
"duration": 11.4,
152+
"channels": 1,
153+
"models": ["nova-3-general"],
154+
"model_info": {
155+
"nova-3-general": {
156+
"name": "nova-3-general",
157+
"version": "2024-12-18.0",
158+
"arch": "nova-3"
159+
}
160+
}
161+
},
162+
"results": {
163+
"channels": [
164+
{
165+
"alternatives": [
166+
{
167+
"transcript": "He doesn't belong to you, and I don't see how you have anything to do with what is be his power, if he possess only that from this stage to you.",
168+
"confidence": 0.9921875,
169+
"words": [
170+
{
171+
"word": "he",
172+
"start": 0.32,
173+
"end": 0.4,
174+
"confidence": 0.9921875,
175+
"speaker": 0,
176+
"punctuated_word": "He"
177+
}
178+
]
179+
}
180+
]
181+
}
182+
]
183+
}
184+
}
185+
}
186+
```
187+
{% endcode %}
188+
189+
</details>

‎docs/api-references/speech-models/speech-to-text/deepgram/nova-3-medical.md‎

Lines changed: 189 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,189 @@
1+
---
2+
layout:
3+
width: wide
4+
title:
5+
visible: true
6+
description:
7+
visible: true
8+
tableOfContents:
9+
visible: true
10+
outline:
11+
visible: true
12+
pagination:
13+
visible: true
14+
metadata:
15+
visible: true
16+
tags:
17+
visible: true
18+
actions:
19+
visible: true
20+
---
21+
22+
# Nova 3 Medical
23+
24+
{% columns %}
25+
{% column width="66.66666666666666%" %}
26+
{% hint style="info" %}
27+
This documentation is valid for the following list of our models:
28+
29+
* `nova-3-medical`
30+
{% endhint %}
31+
{% endcolumn %}
32+
33+
{% column width="33.33333333333334%" %}
34+
<a href="https://aimlapi.com/app/deepgram/nova-3-medical" class="button primary">Try in Playground</a>
35+
{% endcolumn %}
36+
{% endcolumns %}
37+
38+
Nova-3 Medical — speech-to-text model from Deepgram fine-tuned for clinical terminology, healthcare audio, and medical transcription workflows. Accurately recognizes drug names, diagnoses, procedures, and medical jargon in real-time and batch transcription.
39+
40+
{% hint style="success" %}
41+
This model uses per-minute billing. The cost of audio transcription is based on the duration of the input audio file.
42+
{% endhint %}
43+
44+
## Setup your API Key
45+
46+
If you don't have an API key for the AI/ML API yet, feel free to use our [Quickstart guide](https://docs.aimlapi.com/quickstart/setting-up).
47+
48+
## API Schemas
49+
50+
{% openapi-operation spec="deepgram-nova-3-medical" path="/v1/stt/create" method="post" %}
51+
[OpenAPI deepgram-nova-3-medical](https://raw.githubusercontent.com/aimlapi/api-docs/refs/heads/main/docs/api-references/speech-models/Deepgram/nova-3-medical.json)
52+
{% endopenapi-operation %}
53+
54+
## Quick Example: Processing a Speech Audio File via URL
55+
56+
Let's transcribe the following audio fragment:
57+
58+
{% embed url="https://drive.google.com/file/d/1ZN-28NUbK1TXHt6oEPj42zUJCv82e9L4/view?usp=sharing" %}
59+
60+
{% code overflow="wrap" %}
61+
```python
62+
import time
63+
import requests
64+
import json
65+
66+
base_url = "https://api.aimlapi.com/v1"
67+
# Insert your AIML API Key instead of <YOUR_AIMLAPI_KEY>:
68+
api_key = "<YOUR_AIMLAPI_KEY>"
69+
70+
# Creating and sending a speech-to-text conversion task to the server
71+
def create_stt():
72+
url = f"{base_url}/stt/create"
73+
headers = {
74+
"Authorization": f"Bearer {api_key}",
75+
}
76+
77+
data = {
78+
"model": "nova-3-medical",
79+
"url": "https://audio-samples.github.io/samples/mp3/blizzard_primed/sample-0.mp3"
80+
}
81+
82+
response = requests.post(url, json=data, headers=headers)
83+
84+
if response.status_code >= 400:
85+
print(f"Error: {response.status_code} - {response.text}")
86+
else:
87+
response_data = response.json()
88+
print(response_data)
89+
return response_data
90+
91+
# Requesting the result of the task from the server using the generation_id
92+
def get_stt(gen_id):
93+
url = f"{base_url}/stt/{gen_id}"
94+
headers = {
95+
"Authorization": f"Bearer {api_key}",
96+
}
97+
response = requests.get(url, headers=headers)
98+
return response.json()
99+
100+
# First, start the generation, then repeatedly request the result from the server every 10 seconds.
101+
def main():
102+
stt_response = create_stt()
103+
gen_id = stt_response.get("generation_id")
104+
105+
if gen_id:
106+
start_time = time.time()
107+
timeout = 600
108+
while time.time() - start_time < timeout:
109+
response_data = get_stt(gen_id)
110+
111+
if response_data is None:
112+
print("Error: No response from API")
113+
break
114+
115+
status = response_data.get("status")
116+
117+
if status == "waiting" or status == "active":
118+
print("Still waiting... Checking again in 10 seconds.")
119+
time.sleep(10)
120+
else:
121+
print("Processing complete:\n", response_data["result"]["results"]["channels"][0]["alternatives"][0]["transcript"])
122+
123+
# Uncomment the line below to print the entire result object with all service data
124+
# print("Processing complete:\n", json.dumps(response_data["result"], indent=2, ensure_ascii=False))
125+
return response_data
126+
127+
print("Timeout reached. Stopping.")
128+
return None
129+
130+
131+
if __name__ == "__main__":
132+
main()
133+
```
134+
{% endcode %}
135+
136+
<details>
137+
138+
<summary>Response</summary>
139+
140+
{% code overflow="wrap" %}
141+
```json5
142+
{'generation_id': 'c3d4e5f6-a7b8-9012-cdef-123456789012'}
143+
Still waiting... Checking again in 10 seconds.
144+
Processing complete:
145+
{
146+
"status": "completed",
147+
"result": {
148+
"metadata": {
149+
"request_id": "c3d4e5f6-a7b8-9012-cdef-123456789012",
150+
"created": "2025-06-04T12:00:00.000Z",
151+
"duration": 11.4,
152+
"channels": 1,
153+
"models": ["nova-3-medical"],
154+
"model_info": {
155+
"nova-3-medical": {
156+
"name": "nova-3-medical",
157+
"version": "2024-12-18.0",
158+
"arch": "nova-3"
159+
}
160+
}
161+
},
162+
"results": {
163+
"channels": [
164+
{
165+
"alternatives": [
166+
{
167+
"transcript": "He doesn't belong to you, and I don't see how you have anything to do with what is be his power, if he possess only that from this stage to you.",
168+
"confidence": 0.9882813,
169+
"words": [
170+
{
171+
"word": "he",
172+
"start": 0.32,
173+
"end": 0.4,
174+
"confidence": 0.9882813,
175+
"speaker": 0,
176+
"punctuated_word": "He"
177+
}
178+
]
179+
}
180+
]
181+
}
182+
]
183+
}
184+
}
185+
}
186+
```
187+
{% endcode %}
188+
189+
</details>

0 commit comments

Comments
 (0)