SQL Server 2017

By using db<>fiddle, you agree to license everything you submit by Creative Commons CC0.

create table tab (
id int, tokenNo varchar(10), tokenType varchar(10), generatedAt datetime

)

insert into tab values 
(249,'I1','W','2019-11-30 11:07:08.083'),
(250,'I2','W','2019-11-30 11:07:25.230'),
(251,'O1','W','2019-11-30 11:07:35.350'),
(252,'D1','W','2019-11-30 11:58:00.643'),
(253,'O2','W','2019-11-30 11:59:13.370'),
(254,'I3','W','2019-11-30 12:02:59.000'),
(255,'D2','W','2019-11-30 12:04:15.617')

DECLARE @STARTDATE DATETIME = '2019-11-30 07:00:00.000'
DECLARE @ENDDATE DATETIME = '2019-11-30 22:00:00.000'

;WITH FramesCTE AS
(
    SELECT @STARTDATE                 AS [start], 
           DATEADD(hh,3,@STARTDATE)   AS [end], 
           FORMAT(@STARTDATE,'hh tt') AS TimeFrame
    UNION ALL 
    SELECT DATEADD(hh,3,[start]), 
           DATEADD(hh,3,[end]), 
           FORMAT([end],'hh tt')
    FROM   FramesCTE 
    WHERE  [start] < @ENDDATE
)

SELECT TimeFrame, 
       COUNT(generatedAt) AS [Count]
FROM   FramesCTE
LEFT JOIN tab
     ON generatedAt >= [start] 
     AND generatedAt < [end]
GROUP BY TimeFrame

TimeFrame	Count
01 PM	0
04 PM	0
07 AM	0
07 PM	0
10 AM	7
10 PM	0

Warning: Null value is eliminated by an aggregate or other SET operation.